Iron fitting modeling system based on digital twinning

Abstract

The present application relates to the technical field of intelligent manufacturing and remanufacturing based on digital twinning, in particular to a kind of iron fitting modeling system based on digital twinning, comprising: data acquisition module, for obtaining the design specification geometric value corresponding to key geometric feature, design tolerance and functional importance weight;Deviation quantification module is used to determine the total deviation index representing the degree of geometric deviation;Correction value calculation module is used to calculate the final correction geometric value;Model generation module is used to generate idealized digital twinning model;Closed-loop verification module is used to output manufacturing data package or artificial alarm, the present application converts pure geometric measurement problem into functional evaluation problem closely coupled with design intent and functional requirement, realizes the objective, scientific quantitative evaluation to physical entity function compliance degree.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing and remanufacturing technology based on digital twins, specifically to a modeling system for iron outfitting components based on digital twins. Background Technology

[0002] In the field of repair and remanufacturing of in-service equipment components, reverse modeling of key components such as iron outfitting parts is the foundation for generating remanufacturing data. However, due to long-term service, in-service iron outfitting parts inevitably experience wear, deformation, and other geometric deviations in their physical structure. These deviations result in significant differences between the physical geometry and its original design specifications, posing a technical challenge to generating an idealized digital twin model that can be directly used for remanufacturing.

[0003] Traditional reverse modeling techniques primarily focus on accurately reproducing the current geometry of a physical entity, but generally lack effective mechanisms for assessing the functional impact of geometric deviations. Existing methods typically only list a large number of isolated dimensional deviation data, heavily relying on engineers' subjective judgment based on personal experience, making it difficult to reach unified and objective evaluation conclusions. Furthermore, the correction of out-of-tolerance features often employs rigid strategies such as completely replacing them with design values ​​or accepting all scanned values, lacking intelligent means for adaptive and refined correction based on the severity of deviations and functional importance. This current state of technology results in the reverse modeling process failing to guarantee the functional compliance of the model, exhibiting low automation, and struggling to efficiently generate idealized models that are both faithful to the physical prototype and strictly adhere to the design intent.

[0004] Therefore, how to establish a system and method that can deeply integrate physical measurement geometry with design semantics, realize automated quantitative evaluation and intelligent correction of geometric deviations of in-service iron outfitting components, and ultimately generate an idealized digital twin model that can directly guide remanufacturing is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention discloses a modeling system for iron outfitting components based on digital twins. Specifically, the technical solution of this invention is as follows:

[0006] A digital twin-based iron outfitting modeling system includes:

[0007] The data acquisition module is used to obtain the scanning measurement geometric values ​​of key geometric features in the target iron outfitting, and to obtain the design specification geometric values, design tolerances and functional importance weights corresponding to the key geometric features from the preset manufacturing semantic knowledge base.

[0008] The deviation quantification module is used to determine the total deviation index, which characterizes the degree of geometric deviation, based on the geometric values ​​of scanning measurements, geometric values ​​of design specifications, design tolerances, and functional importance weights.

[0009] The correction value calculation module is used to determine the semantic fusion weights based on the total deviation index, and to calculate the final correction geometric values ​​based on the semantic fusion weights, scanning measurement geometric values ​​and design specification geometric values.

[0010] The model generation module is used to integrate the final corrected geometric values ​​to generate an idealized digital twin model;

[0011] The closed-loop verification module is used to calculate the final deviation index of the idealized digital twin model, and output manufacturing data packets or manual alarms based on the final deviation index and the preset acceptance threshold.

[0012] Preferably, the data acquisition module includes:

[0013] The point cloud acquisition unit is used to acquire the original geometric dataset of the target iron outfitting using 3D scanning technology;

[0014] The feature extraction unit is used to automatically identify and extract key geometric features from the original geometric dataset to obtain scan measurement geometric values;

[0015] The semantic query unit is used to query and retrieve design specification geometric values, design tolerances, and functional importance weights associated with key geometric feature mappings from the manufacturing semantic knowledge base.

[0016] Preferably, the deviation quantification module determines the total deviation index, including:

[0017] Calculate the absolute deviation between the scanned geometric value and the design specification geometric value of each key geometric feature, and divide it by the design tolerance corresponding to that feature to obtain the normalized individual deviation.

[0018] The total deviation index is determined by multiplying the individual deviation by the functional importance weight corresponding to the feature and summing the weighted results of all key geometric features.

[0019] Preferably, the system also includes:

[0020] The deviation determination module is used to compare the total deviation index with the preset first-level correction activation threshold to determine whether the geometric state of the key geometric features is compliant, first-level disqualified, or second-level disqualified.

[0021] Preferably, the correction value calculation module determines the semantic fusion weights, including:

[0022] Call the total deviation index and the steepness coefficient and correction center point, which are preset correction control parameters;

[0023] Calculate the relative deviation of the total deviation index from the correction center point;

[0024] The relative deviation is input into a normalized logistic function controlled by a steepness coefficient to generate semantic fusion weights.

[0025] Preferably, the correction value calculation module calculates the final corrected geometric value, including:

[0026] A linear interpolation model is used to perform weighted calculations on the geometric values ​​of the design specifications and the geometric values ​​of the scanning measurements;

[0027] The weights for the design specification geometric values ​​are semantic fusion weights, and the weights for the scan measurement geometric values ​​are the difference between 1 and the semantic fusion weights.

[0028] Preferably, the model generation module generates an idealized digital twin model, including:

[0029] Receive the geometric state output by the deviation determination module;

[0030] Filter out key geometric features that are disqualified at level one or level two;

[0031] The final corrected geometric values ​​corresponding to all the selected key geometric features are integrated to generate an idealized digital twin model.

[0032] Preferably, the closed-loop verification module includes:

[0033] The final inspection execution unit is used to take the idealized digital twin model as input, call the deviation quantization module, and recalculate the final deviation index.

[0034] The result determination unit is used to compare the final deviation index with the preset acceptance threshold and generate the verification result.

[0035] The output control unit is used to generate a manufacturing data packet when the verification result is successful, and to issue a manual alarm when the verification result is unsuccessful.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. This invention integrates multiple independent geometric deviations into a single total deviation index for comprehensive evaluation by introducing design tolerance and functional importance weights. This overcomes the shortcomings of traditional reverse modeling, which relies on the subjective experience and judgment of engineers, and transforms the purely geometric measurement problem into a functional evaluation problem closely coupled with design intent and functional requirements, thus realizing an objective and scientific quantitative evaluation of the functional conformity of physical entities.

[0038] 2. This invention proposes a dynamic correction mechanism based on semantic fusion weights. The system automatically calculates correction weights according to the severity of the quantified deviation and smoothly and continuously fuses the scanned measurement values ​​and design specification values. This avoids the rigid, either-or correction choices in traditional methods, and can retain reasonable morphological features of the entity to the maximum extent while correcting functional deviations, so that the generated model has both accuracy and fidelity.

[0039] 3. This invention achieves automatic classification of geometric states by establishing a deviation judgment module; during model generation, targeted correction is performed only on key geometric features judged as disqualifying, while compliant parts are retained. This approach greatly improves the efficiency of model processing, avoids unnecessary data operations on the overall model, and achieves a balance between accuracy, automation, and efficiency in correction.

[0040] 4. This invention constructs an automated closed-loop process from deviation quantification and model correction to final verification. By performing final deviation index calculation and threshold comparison on the corrected idealized digital twin model, the quality of the output model is ensured to meet preset standards. After verification, a manufacturing data package containing a 3D model and processing code can be directly generated, opening up the entire process from physical scanning to digital manufacturing and providing a rigorously verified high-quality data source for downstream remanufacturing. Attached Figure Description

[0041] The present invention will be further explained below with reference to the accompanying drawings and embodiments:

[0042] Figure 1 This is a flowchart of the system of the present invention.

[0043] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0044] Example 1

[0045] Please see Figure 1 A digital twin-based iron outfitting modeling system includes:

[0046] The data acquisition module is used to obtain the scanning measurement geometric values ​​of key geometric features in the target iron outfitting, and to obtain the design specification geometric values, design tolerances and functional importance weights corresponding to the key geometric features from the preset manufacturing semantic knowledge base.

[0047] The deviation quantification module is used to determine the total deviation index, which characterizes the degree of geometric deviation, based on the geometric values ​​of scanning measurements, geometric values ​​of design specifications, design tolerances, and functional importance weights.

[0048] The correction value calculation module is used to determine the semantic fusion weights based on the total deviation index, and to calculate the final correction geometric values ​​based on the semantic fusion weights, scanning measurement geometric values ​​and design specification geometric values.

[0049] The model generation module is used to integrate the final corrected geometric values ​​to generate an idealized digital twin model;

[0050] The closed-loop verification module is used to calculate the final deviation index of the idealized digital twin model, and output manufacturing data packets or manual alarms based on the final deviation index and the preset acceptance threshold.

[0051] This embodiment provides a modeling system for iron outfitting based on digital twins. The system aims to solve the technical problem that geometric deviations caused by factors such as wear and deformation of physical entities are difficult to be effectively evaluated and automatically corrected during the reverse modeling of in-service iron outfittings, thus making it impossible to generate an ideal digital twin model that can be directly used for remanufacturing.

[0052] The system includes: a data acquisition module, a deviation quantification module, a correction value calculation module, a model generation module, and a closed-loop verification module;

[0053] The data acquisition module aims to comprehensively and accurately acquire all geometric and semantic data required for subsequent evaluation and correction. In this embodiment, the data acquisition module uses non-contact measurement technologies such as 3D laser scanning to acquire high-density point cloud data of the target iron outfitting surface, forming a raw geometric dataset reflecting its current physical state. The module's built-in algorithm automatically identifies and extracts predefined key geometric features from the dataset, such as planes, holes, and edge contours, and accurately calculates the actual measured geometric parameters of each feature, which is the scanned measurement geometric value. It refers to the actual dimensional parameters that characterize a specific geometric feature on a physical iron outfitting, obtained through physical measuring equipment, and its source is three-dimensional scanning point cloud data;

[0054] At the same time, the data acquisition module accesses a pre-set manufacturing semantic knowledge base; this knowledge base stores design specification geometric values ​​that correspond one-to-one with the aforementioned key geometric features. Design tolerances and functional importance weight It refers to the ideal theoretical dimension set for a certain geometric feature based on product design drawings or relevant manufacturing standards, and its source is the manufacturing semantic knowledge base; This refers to the amount of variation in the actual size of a geometric feature relative to its theoretical size, provided that functional requirements are met. The source of this variation is the semantic knowledge base. This refers to a dimensionless coefficient preset based on the functional importance of a certain geometric feature in the entire iron outfitting, such as whether it is an assembly datum or a major load-bearing part. The source of this coefficient is the manufacturing semantic knowledge base. In this way, the module accurately associates and pairs the actual geometry of the physical entity with the ideal semantics of the engineering design, providing complete data input for subsequent quantitative analysis.

[0055] The purpose of the deviation quantification module is to integrate multiple discrete geometric deviation values ​​into a single index that can macroscopically evaluate the overall functional compliance. In this embodiment, the deviation quantification module receives data provided by the data acquisition module. It first calculates the absolute value of the difference between the actual measured value and the design specification value for each key geometric feature. Then use the design tolerance of this feature. The deviation is normalized to obtain a dimensionless one-way bias. The technical value of this normalization step lies in its elimination of the scaling effect caused by different features, such as length and angle, and their varying tolerances, making the deviations of different features comparable. Subsequently, the module multiplies the normalized one-way bias of each feature by its corresponding functional importance weight. Finally, the total deviation index, which characterizes the overall degree of geometric deviation, is determined by summing all the weighted individual deviations. The calculation of this index follows the mathematical model:

[0056]

[0057] in: The total deviation index, a dimensionless value, is the final output of this module.

[0058] The total number of key geometric features to be evaluated is an integer, and its value is determined by the number of features identified by the data acquisition module.

[0059] No. The functional importance weights of each feature are dimensionless floating-point numbers in the range [0,1], and their source is the semantic knowledge base.

[0060] No. The scanning measurement geometric values ​​of each feature are physical quantities such as length or angle, and their source is the calculation results of the data acquisition module;

[0061] No. The design specification geometric values ​​of each feature, and They have the same units of measurement, and their source is the creation of semantic knowledge bases;

[0062] The design tolerance of the feature, and They have the same units of measurement, and their source is the creation of semantic knowledge bases;

[0063] Through this design, The value not only reflects the absolute magnitude of the deviation, but also incorporates considerations of functional importance, making the evaluation results more closely related to engineering practice;

[0064] The correction value calculation module aims to intelligently calculate the target geometric value to which each disqualifying feature should be corrected, based on the severity of the deviation. In this embodiment, the correction value calculation module uses the total deviation index output by the deviation quantification module. Determine a semantic fusion weight The calculation was designed as a normalized logistic function to ensure the smoothness and controllability of the correction process; subsequently, the module was based on this... Weights, applied through a linear interpolation model, are used to measure the geometric values ​​of the scan. Geometric values ​​in accordance with design specifications Perform fusion calculations to obtain the final corrected geometric values. This refers to the ideal geometric value, calculated by this system, used to replace the original disqualifying features, which conforms to design specifications while retaining some of the original form; its calculation logic is that when the deviation is severe... When the value is high, Approaching 1, making Closer to the design specification value Forced correction is implemented; when the deviation is slight, Approaching 0, making Closer to scan measurement value To preserve more original morphological information;

[0065] The model generation module aims to update the 3D model using calculated correction values. In this embodiment, the model generation module iterates through all key geometric features determined to require correction and replaces their original geometric parameters with the final corrected geometric values ​​calculated by the correction value calculation module. After replacing all disqualifying features, the module reconstructs the topological relationships and solves the geometric constraints of the entire model, ultimately generating a brand-new idealized digital twin model whose geometric state fully conforms to the design intent. This process involves calling the model's parametric solver, recalculating the geometric constraints associated with the corrected features, and updating the model's boundary representation topology to ensure that the modified features can be seamlessly and correctly integrated with the rest of the model.

[0066] The closed-loop verification module aims to perform a final quality check on the generated idealized digital twin model to ensure it meets the output standards. In this embodiment, the closed-loop verification module uses the idealized digital twin model output by the model generation module as new input, and again invokes all functions of the deviation quantification module to perform a complete deviation evaluation, calculating a final deviation index. This refers to the result obtained after further quantitatively evaluating the deviation of the corrected idealized digital twin model; its purpose is to verify the effectiveness of the correction process. The module will... With a preset acceptance threshold Compare; This is a dimensionless constant slightly greater than 1, designed to tolerate minute residual deviations caused by the precision of computer floating-point operations or the convergence of algorithms. Its value can be determined by referring to the industry's general requirements for CNC machining precision. If... Not greater than If the verification is successful, the system will automatically output a manufacturing data package containing the idealized digital twin model file and the corresponding CNC machining code; otherwise, if... If the verification fails, the system will mark the model and issue a manual alert, prompting technical personnel to intervene and check.

[0067] This embodiment constructs a complete technical process from physical entity scanning, functional deviation quantitative assessment, semantic-driven automated correction to closed-loop verification through the collaborative work of the above modules. It overcomes the shortcomings of traditional reverse modeling, which can only reproduce geometric shapes but cannot guarantee functional compliance. It can automatically and efficiently generate idealized digital twin models that are both faithful to the physical prototype and strictly follow the design intent, and directly output production data that can be used for remanufacturing, which greatly improves the efficiency and quality of in-service equipment parts repair and remanufacturing.

[0068] Example 2:

[0069] The data acquisition module includes:

[0070] The point cloud acquisition unit is used to acquire the original geometric dataset of the target iron outfitting using 3D scanning technology;

[0071] The feature extraction unit is used to automatically identify and extract key geometric features from the original geometric dataset to obtain scan measurement geometric values;

[0072] The semantic query unit is used to query and retrieve design specification geometric values, design tolerances, and functional importance weights associated with key geometric feature mappings from the manufacturing semantic knowledge base.

[0073] This embodiment is a specific implementation of the data acquisition module described in Embodiment 1, which aims to further improve the automation level of data acquisition and the accuracy of semantic association;

[0074] In this specific implementation, the data acquisition module is divided into three collaborative units: a point cloud acquisition unit, a feature extraction unit, and a semantic query unit.

[0075] The point cloud acquisition unit aims to acquire the original geometric shape data of the physical iron outfitting. In this embodiment, the unit uses three-dimensional scanning technology, such as a handheld or fixed laser scanner, to perform an all-round scan of the target iron outfitting in order to obtain an original geometric dataset that can accurately characterize its surface morphology. This dataset usually exists in the form of a high-density point cloud.

[0076] The feature extraction unit aims to extract engineering-significant geometric structures from massive, unordered point cloud data. In this embodiment, the unit automatically identifies and extracts key geometric features from the original geometric dataset. It can use algorithms such as RANSAC to automatically segment the point cloud and fit geometric primitives such as planes, cylindrical surfaces, circular holes, and edge lines. After identification, the unit accurately calculates the geometric parameters of these features, such as the normal vector of the plane, the diameter and axis of the hole, and the curvature of the edge line. These parameters constitute the scanning measurement geometric values.

[0077] The semantic query unit aims to assign engineering design-level meanings and constraints to geometric features extracted from the physical world. In this embodiment, the unit queries and retrieves design specification geometric values, design tolerances, and functional importance weights associated with key geometric features from a manufacturing semantic knowledge base. When the feature extraction unit identifies a feature, such as a hole with a diameter of 50.2 mm, the semantic query unit uses the feature's type, approximate location, and size as indexes to match it in the knowledge base, finding its corresponding design entry, such as an assembly hole marked Ø50H7 on a design drawing, thereby accurately obtaining its design specification geometric value. Design tolerances and preset functional importance weights

[0078] By specifying the data acquisition module into the three units mentioned above, this invention realizes a fully automated processing flow from raw point clouds to structured and semantic data. Compared with general data acquisition, this structured approach automatically extracts features through algorithms and accurately maps them to a semantic library, avoiding the inefficiency and subjective errors of manual measurement. This significantly improves the efficiency and reliability of the data preparation stage and lays a solid foundation for subsequent accurate deviation quantification.

[0079] Example 3:

[0080] The deviation quantification module determines the total deviation index, including:

[0081] Calculate the absolute deviation between the scanned geometric value and the design specification geometric value of each key geometric feature, and divide it by the design tolerance corresponding to that feature to obtain the normalized individual deviation.

[0082] The total deviation index is determined by multiplying the individual deviation by the functional importance weight corresponding to the feature and summing the weighted results of all key geometric features.

[0083] This embodiment is a specific algorithm implementation of the process of determining the total deviation index by the deviation quantification module described in Embodiment 1, and aims to elaborate on the construction logic of this core index in detail;

[0084] In this specific implementation, the deviation quantization module performs the following calculation steps to determine the total deviation index.

[0085] Calculate the scan measurement geometric values ​​for each key geometric feature. Geometric values ​​in accordance with design specifications The absolute deviation, divided by the design tolerance corresponding to that feature. We obtain the normalized one-way bias; specifically, for the first... Features, module execution The calculation; this step is not simply to measure dimensional differences, but to measure the severity of the difference within its permissible range of variation, i.e., tolerance; a feature with a deviation of 0.1 mm, if its tolerance is 0.05 mm, is far more serious than another feature with a deviation of 0.2 mm but a tolerance of 0.5 mm; this normalization process puts the deviations of all features on a uniform scale that is directly related to functional requirements for evaluation;

[0086] Multiply the individual bias obtained in the preceding steps by the functional importance weight corresponding to that feature. The weighted results of all key geometric features are then summed to determine the total deviation index. Specifically, module execution The calculation; this step introduces the concept of importance in engineering semantics; for example, a flatness deviation of an assembly reference surface has a far greater impact on the final product's functionality than a radius deviation of a common decorative fillet; by multiplying by The system can amplify the impact of deviations in important features on the overall evaluation results, while reducing the interference of deviations in minor features, thus improving the final total deviation index. It can more realistically and accurately reflect the degree to which a physical entity conforms to its design intent at the functional level;

[0087] This specific method for determining the total deviation index creatively transforms a purely geometric measurement problem into a functional evaluation problem closely coupled with design intent and functional requirements through two core steps: tolerance normalization and weighting. Compared with the existing approach of listing a large number of isolated dimensional deviations and requiring engineers to make judgments based on experience, this embodiment provides a single evaluation index that is objective, quantitative, and has clear engineering guidance significance, which greatly improves the scientificity and automation level of deviation evaluation.

[0088] Example 4:

[0089] A digital twin-based iron outfitting modeling system also includes:

[0090] The deviation determination module is used to compare the total deviation index with the preset first-level correction activation threshold to determine whether the geometric state of the key geometric features is compliant, first-level disqualified, or second-level disqualified.

[0091] This embodiment adds a deviation determination module to the system described in Embodiment 1. Its purpose is to analyze the total deviation index calculated by the deviation quantification module and output a handling conclusion with clear guiding significance.

[0092] In this embodiment, the system further includes a deviation determination module; this module determines the total deviation index. Compared with the preset first-level correction activation threshold Comparisons are made to determine the geometric state of key geometric features; a first-level correction activation threshold is applied. This refers to a pre-defined dimensionless critical value used to distinguish the severity levels of deviations; its function is to transform continuous deviation indices into discrete, operationally guideable status levels; its setting logic is based on risk assessment and correction cost considerations, for example, for critical load-bearing components with extremely high safety requirements. A smaller value, such as 1.5, will be used to ensure that even minor deviations are flagged promptly; while for general components, A larger value, such as 3.0, can be selected;

[0093] Its decision logic is set as follows:

[0094] Compliance Status When the calculated total deviation index When the value is less than or equal to 1, it is considered to be in compliance status; this indicates that even if there are measurement deviations of multiple features, after comprehensively considering their respective tolerances and functional importance weights, the overall geometric state is still within the design-allowed range and no correction is required.

[0095] First-class disqualification Value greater than 1 but less than or equal to The system has been classified as a Level 1 failure. This indicates that the geometric deviation has exceeded the range of the overall weighted tolerance. Although it has not yet reached a critical level, it has already had a potential impact on the intended function, and the system will trigger the subsequent correction process.

[0096] Second-degree disqualification The system is classified as a Level 2 disqualification; this indicates a very serious geometric deviation that could lead to assembly failure or major functional defects, requiring mandatory and more significant corrections. By adding a deviation determination module, this system further automates decision-making based on the quantification of deviations; it uses a continuous numerical... The process is transformed into clear engineering instructions such as compliance, first-level disqualification, and second-level disqualification, providing a clear basis for action for the subsequent model generation module. This means identifying which features need to be corrected and the urgency of the correction, avoiding the subjectivity of manual interpretation and decision-making, and making the entire modeling and correction process more intelligent and standardized.

[0097] Example 5:

[0098] The correction value calculation module determines the semantic fusion weights, including:

[0099] Call the total deviation index and the steepness coefficient and correction center point, which are preset correction control parameters;

[0100] Calculate the relative deviation of the total deviation index from the correction center point;

[0101] The relative deviation is input into a normalized logistic function controlled by a steepness coefficient to generate semantic fusion weights.

[0102] This embodiment determines the semantic fusion weights using the correction value calculation module described in Embodiment 1. This is a specific algorithmic implementation of the process, which aims to explain in detail how to generate a stable and controllable correction strategy based on the deviation index.

[0103] In this specific implementation, the correction value calculation module performs the following steps to determine the semantic fusion weights.

[0104] Call the total deviation index Steepness coefficient, which is a preset correction control parameter and correction center point Steepness coefficient This refers to a dimensionless parameter that controls the weights. Follow The rate of change, i.e. how quickly the correction strategy transitions from maintaining the status quo to forcibly aligning with the standard; The larger the value, the more drastic the transition; correct the center point. This refers to a dimensionless parameter that represents the center of symmetry of the logistic function curve; in engineering, it defines the weights. The deviation index value of 0.5 represents the equilibrium point of the correction strategy. Both parameters are preset in the system by technicians according to correction requirements. For example, if it is desired to apply a high-weight correction quickly once the deviation exceeds the tolerance, a larger value is set.

[0105] Calculate the total deviation index Relative to the correction center point Relative deviation; module execution The calculation; the technical motivation of this design is to achieve normalization in order to eliminate The influence of the numerical scale itself on the correction behavior; regardless of Whether it is set to 1.5 or 5.0, when achieve twice as much At that time, the relative deviation is always 1; this makes The physical meaning of the value becomes clear and stable;

[0106] Input the relative deviation calculated in the previous step into the steepness coefficient. In the normalized logistic function of control, semantic fusion weights are generated. The specific mathematical expression of this function is:

[0107]

[0108] Using this model, when much smaller The exponent term is a large positive number. Approaching 0; The exponent term is 0. Equals 0.5; when Much larger When the exponent term is a large negative number, Approaching 1; the entire process transitions smoothly, and its transition behavior can be... and Perform precise engineering settings;

[0109] This determines the semantic fusion weights. The specific method involves introducing a normalized logistic function and providing... Two adjustment parameters with clear engineering significance solve the problem of difficulty in quantifying and controlling the correction intensity in traditional correction methods. This makes the generation process of correction strategy not only automatic, but also predictable, configurable and highly robust, ensuring the feasibility and consistency of the technical solution in different engineering scenarios, and greatly improving the controllability and robustness of the correction strategy.

[0110] Example 6:

[0111] The correction value calculation module calculates the final corrected geometric values, including:

[0112] A linear interpolation model is used to perform weighted calculations on the geometric values ​​of the design specifications and the geometric values ​​of the scanning measurements;

[0113] The weights for the design specification geometric values ​​are semantic fusion weights, and the weights for the scan measurement geometric values ​​are the difference between 1 and the semantic fusion weights.

[0114] This embodiment calculates the final corrected geometric value using the correction value calculation module described in Embodiment 1. A specific algorithmic implementation of the process, aiming to illustrate how to apply the correction strategy, i.e., the weights. It is applied to specific correction operations; in this specific implementation, the correction value calculation module determines the semantic fusion weights. Next, the following steps are performed to calculate the final corrected geometry values.

[0115] A linear interpolation model is used to evaluate the geometric values ​​of the design specifications. With scanning measurement geometry Weighted calculations are performed; the linear interpolation model is chosen because it is simple in form, computationally stable, and has clear physical meaning. It represents a trade-off between the ideal standard and the current physical state in the final correction result.

[0116] Design Specification Geometric Values The weights are semantic fusion weights. Scanning measurement geometric values The weight is 1 and the semantic fusion weight The difference; the underlying logic of this weight allocation lies in, The value is determined by the severity of the deviation. Therefore, when the deviation is severe, When the value approaches 1, the model tends to adopt the value that represents the design intent. When the deviation is slight Approaching 0, The value approaches 1, indicating that the model tends to adopt a value that represents the current physical state. The entire calculation process is precisely expressed by the following mathematical model:

[0117]

[0118] in: The final corrected geometric value has the same physical dimensions as... Maintain consistency;

[0119] Dimensionless semantic fusion weights calculated by the method in Example 5;

[0120] The geometric values ​​for the design specifications are derived from the data acquisition module.

[0121] The scanning measurement of geometric values ​​originates from the data acquisition module;

[0122] By adopting this based The present invention implements a dynamic and adaptive geometric correction mechanism based on the weighted linear interpolation model. It avoids the rigid binary choice in traditional methods, which either accept the scan value entirely or completely replace it with the design value. Instead, it performs a smooth and continuous fusion according to the severity of the problem. The final corrected geometric value generated by this method not only corrects unacceptable functional deviations, but also retains the reasonable morphological features of the physical entity's non-functional influences to the greatest extent, making the final digital twin model more realistic and effective.

[0123] Example 7:

[0124] The model generation module generates idealized digital twin models, including:

[0125] Receive the geometric state output by the deviation determination module;

[0126] Filter out key geometric features that are disqualified at level one or level two;

[0127] The final corrected geometric values ​​corresponding to all the selected key geometric features are integrated to generate an idealized digital twin model.

[0128] This embodiment is a detailed explanation of the process by which the model generation module generates an idealized digital twin model, based on embodiment 4. It aims to illustrate how this module works in conjunction with the deviation determination module to achieve targeted and efficient model correction.

[0129] In this implementation, the execution flow of the model generation module is tightly coupled with the output of the deviation determination module:

[0130] The first step is that the model generation module receives the geometric state output by the deviation determination module; as described in the implementation of Example 4, the deviation determination module will determine the geometric state of the iron outfitting as compliant, first-level disqualified or second-level disqualified.

[0131] The second step involves the model generation module filtering out key geometric features that are classified as either Level 1 or Level 2 disqualifying based on the received status information. This filtering step is a prerequisite for targeted correction. For features deemed compliant, the model generation module will directly adopt their original scanned geometric values. Without making any modifications, the geometric parts of the physical entity that already meet the requirements are preserved;

[0132] The third step is for the model generation module to integrate the final corrected geometric values ​​corresponding to all the selected key geometric features. To generate an idealized digital twin model; for each selected first-level or second-level disqualification feature, the model generation module obtains the final corrected geometric value calculated for it from the correction value calculation module. and use this The value replaces the original value of this feature in the model. Value; After all disqualifying features have been replaced, the module performs the final geometric constraint solution and model integration to generate an idealized digital twin model;

[0133] This implementation demonstrates the synergistic effect between the deviation determination module and the model generation module, achieving synergistic efficiency among the modules. It decomposes a global correction task into precise operations on disqualifying features. This targeted correction method not only greatly improves the efficiency of model processing and avoids unnecessary data operations on compliant parts, but more importantly, it ensures that the final generated model retains the original valid information of the physical entity to the maximum extent, avoiding model distortion that may be caused by globally forced alignment design standards, and achieving a balance between the accuracy and fidelity of the correction.

[0134] Example 8:

[0135] The closed-loop verification module includes:

[0136] The final inspection execution unit is used to take the idealized digital twin model as input, call the deviation quantization module, and recalculate the final deviation index.

[0137] The result determination unit is used to compare the final deviation index with the preset acceptance threshold and generate the verification result.

[0138] The output control unit is used to generate a manufacturing data packet when the verification result is successful, and to issue a manual alarm when the verification result is unsuccessful.

[0139] This embodiment is a specific structured implementation of the closed-loop verification module described in Embodiment 1, aiming to explain in detail how to ensure the quality and reliability of the final output model through an automated final inspection process;

[0140] In this specific implementation, the closed-loop verification module is internally divided into three functionally sequential units: the final inspection execution unit, the result determination unit, and the output control unit.

[0141] The final inspection execution unit aims to perform an independent and objective quality assessment of the revised model. In this embodiment, the unit takes the idealized digital twin model as input, calls the deviation quantification module, and recalculates the final deviation index. This process reuses the deviation quantification algorithm and parameters from the previous steps, and uses the same evaluation benchmark to verify the correction results, thus forming a rigorous quality verification closed loop.

[0142] The result determination unit aims to make a clear decision of pass or failure based on the final inspection result; in this embodiment, this unit will use the final deviation index calculated by the final inspection execution unit. With the preset acceptance threshold The comparison is performed, and verification results are generated; as mentioned earlier, It is a constant that allows for a machine precision error slightly greater than 1; if Then the unit generates a verified result; if Then a verification failure result will be generated;

[0143] The output control unit is designed to perform subsequent operations based on the ruling result, thus completing the final link from model to manufacturing. In this embodiment, the unit generates a manufacturing data package when the verification result is successful. This data package typically includes a 3D model file in a standard format such as STEP or IGES, as well as G-code or CAM path files that can be directly used on CNC machine tools. When the verification result is unsuccessful, the unit issues a manual alarm. This alarm can be sent to technical personnel through system interface pop-ups, emails, or log records, notifying them that the automated correction of the model has failed to meet the final quality requirements and that manual intervention is needed for diagnosis and processing.

[0144] By specifying the closed-loop verification module into these three units, this invention constructs an automated and standardized final quality inspection process. Compared to the tedious manual dimensional verification required after traditional modeling, this embodiment achieves automated final inspection of the model's functional compliance. This not only greatly improves verification efficiency and eliminates potential oversights during manual inspection, but more importantly, it provides downstream manufacturing processes with a rigorously verified high-quality data source, ensuring the reliability and consistency of the entire process from physical scanning to digital manufacturing. This is a key guarantee for realizing a closed-loop intelligent manufacturing system.

[0145] The above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention; any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A digital twin based modeling system for iron workpieces, characterized in that, The method comprises the following steps: a data acquisition module is used to obtain the scanning measurement geometry value of the key geometric feature in the target iron fitting, and obtain the design specification geometry value, design tolerance and functional importance weight corresponding to the key geometric feature from the preset manufacturing semantic knowledge base; a deviation quantification module is used to determine the total deviation index representing the degree of geometric deviation based on the scanning measurement geometry value, design specification geometry value, design tolerance and functional importance weight; a correction value calculation module is used to determine the semantic fusion weight according to the total deviation index, and calculate the final correction geometry value based on the semantic fusion weight, scanning measurement geometry value and design specification geometry value; the correction value calculation module determines the semantic fusion weight, comprising: calling the total deviation index, the steepness coefficient and the correction center point as the preset correction control parameter; calculating the relative deviation degree of the total deviation index relative to the correction center point; that is, calculating the difference between the total deviation index and the correction center point, and dividing the correction center point to obtain the relative deviation degree; inputting the relative deviation degree into the normalized logistic function controlled by the steepness coefficient to generate the semantic fusion weight; that is, calculating the opposite number of the product of the steepness coefficient and the relative deviation degree, and calculating the reciprocal of the sum of 1 and the natural exponential power with the opposite number as the index to generate the semantic fusion weight; the correction value calculation module calculates the final correction geometry value, comprising: using a linear interpolation model to perform weighted calculation on the design specification geometry value and the scanning measurement geometry value; the weight of the design specification geometry value is the semantic fusion weight, and the weight of the scanning measurement geometry value is the difference between 1 and the semantic fusion weight; a model generation module is used to integrate the final correction geometry value to generate an idealized digital twin model; a closed-loop verification module is used to calculate the final deviation index of the idealized digital twin model, and output the manufacturing data package or manual alarm according to the final deviation index and the preset acceptance threshold.

2. The digital-twin-based modeling system for iron fittings according to claim 1, wherein, The data acquisition module comprises: a point cloud acquisition unit is used to obtain the original geometry data set of the target iron fitting by using a three-dimensional scanning technology; a feature extraction unit is used to automatically identify and extract the key geometric feature from the original geometry data set to obtain the scanning measurement geometry value; a semantic query unit is used to query and obtain the design specification geometry value, design tolerance and functional importance weight associated with the key geometric feature from the manufacturing semantic knowledge base.

3. The digital-twin-based modeling system for iron fittings according to claim 1, wherein, The deviation quantification module determines the total deviation index, comprising: calculating the absolute deviation of the scanning measurement geometry value and the design specification geometry value of each key geometric feature, and dividing the design tolerance corresponding to the feature to obtain the normalized single deviation; multiplying the single deviation by the functional importance weight corresponding to the feature, and summing the weighted results of all key geometric features to determine the total deviation index.

4. The digital-twin-based modeling system for iron fittings according to claim 1, wherein, Further comprising: a deviation determination module is used to compare the total deviation index with the preset first-level correction activation threshold to determine the geometric state of the key geometric feature as a compliant state, a first-level disqualification or a second-level disqualification.

5. The digital twin-based modeling system for iron fittings according to claim 4, wherein, The model generation module generates an idealized digital twin model, comprising: receiving the geometric state output by the deviation determination module; screening out the key geometric features with the geometric state of the first-level disqualification or the second-level disqualification; Integrate all the final modified geometry values corresponding to the screened key geometric features to generate an idealized digital twin model.

6. The digital twin-based modeling system for ironware according to claim 1, wherein, The closed-loop verification module comprises: A final inspection execution unit configured to take the idealized digital twin model as input, call the deviation quantification module, and recalculate the final deviation index; A result determination unit configured to compare the final deviation index with a preset acceptance threshold and generate a verification result; An output control unit configured to generate a manufacturing data package when the verification result is passed, and issue an artificial alarm when the verification result fails.

Images