Progressive multi-point forming process correction method for special-shaped sheet metal parts

Abstract

The application is suitable for the sheet metal technical field, and particularly relates to a progressive multi-point forming process correction method for a special-shaped sheet metal part, which comprises the following steps: determining a plurality of low-risk forming combinations based on blank basic data, process parameter data and equipment state data, determining a suspicious defect area, determining a suspiciously associated influence area based on a multi-dimensional matching result, and obtaining a target defect area set, wherein the target defect area set is determined in the low-risk forming combination, the suspicious defect area is not only screened out, and the suspiciously associated influence area associated according to a cause is screened out, a specific target defect area set and the associated multi-dimensional matching result are obtained, a targeted forming process correction scheme can be generated, and the above method can solve the problems that the traditional process secondary correction is mostly "one-size-fits-all" adjustment, a differentiated scheme is not formulated in combination with the specific cause of the defect area, the correction efficiency is low, and new defects are easily caused.

Description

Technical Field

[0001] This application belongs to the field of sheet metal technology, and in particular relates to a method for modifying the progressive multi-point forming process of irregular sheet metal parts. Background Technology

[0002] Irregularly shaped sheet metal parts are widely used in high-end equipment fields such as aerospace, automobile manufacturing, rail transportation, and construction machinery due to their advantages such as lightweight, high strength, and strong spatial adaptability.

[0003] Traditional secondary correction processes often employ a "one-size-fits-all" approach, failing to develop differentiated solutions based on the specific causes of defects in different areas. This results in low correction efficiency and a high risk of introducing new defects. Because the specific causes of defects are not analyzed (whether it's due to uneven billet properties, unreasonable local process parameters, or localized equipment precision deviations), global parameter adjustments can be ineffective. For example, reducing the overall forming speed to address localized wrinkling can lead to decreased forming efficiency in other areas of the part, and may even cause new problems such as increased springback due to excessively low deformation rates. Conversely, reducing the overall punch pitch to address localized cracking can significantly increase forming time and reduce production efficiency. Summary of the Invention

[0004] This application provides a progressive multi-point forming process correction method for irregularly shaped sheet metal parts, which can solve the problem that traditional secondary correction processes are mostly "one-size-fits-all" adjustments, without taking into account the specific causes of the defect area to formulate differentiated solutions, resulting in low correction efficiency and easy to cause new defects.

[0005] In a first aspect, embodiments of this application provide a method for modifying the progressive multi-point forming process of irregularly shaped sheet metal parts, including: Obtain the forming parameters of the irregularly shaped sheet metal part; wherein, the forming parameters include basic data of the blank, process parameter data and equipment status data; Based on the billet data, the process parameter data, and the equipment status data, several low-risk forming combinations are determined. Based on various low-risk forming combinations, the suspected defect areas of irregularly shaped sheet metal parts are identified; Based on the suspected defect area of ​​the irregular sheet metal part, a multi-dimensional matching result is obtained to determine the suspected associated influence area; wherein, the suspected associated influence area refers to other forming areas that are affected by the defect in the suspected defect area; the multi-dimensional matching result refers to the result obtained by matching the parameter similarity between the local area of ​​the irregular sheet metal part and the suspected defect area from three dimensions: basic data of the blank, process parameter data, and equipment status data. The suspected defect area and the suspected associated influence area of ​​the irregular sheet metal part are merged into a target defect area set; Based on the target defect region set and the multi-dimensional matching results, a forming process correction scheme is output; wherein, the forming process correction scheme is a forming scheme that requires secondary adjustment of forming parameters during the low-risk forming combination forming process of irregular sheet metal parts.

[0006] The technical solutions described in this application embodiment have at least the following technical effects: The progressive multi-point forming process correction method for irregularly shaped sheet metal parts provided in this application embodiment obtains the forming parameters of the irregularly shaped sheet metal parts. These forming parameters include basic blank data, process parameter data, and equipment status data. Based on the basic blank data, process parameter data, and equipment status data, multiple low-risk forming combinations are determined. Based on these multiple low-risk forming combinations, suspected defect areas of the irregularly shaped sheet metal parts are determined. Based on the suspected defect areas of the irregularly shaped sheet metal parts, multi-dimensional matching results are obtained to determine suspected associated influence areas. The suspected associated influence areas represent other forming areas affected by the defect in the suspected defect area. The multi-dimensional matching results refer to the results obtained by matching the parameter similarity between local areas of the irregularly shaped sheet metal parts and suspected defect areas from three dimensions: basic blank data, process parameter data, and equipment status data. The suspected defect areas and suspected associated influence areas of the irregularly shaped sheet metal parts are merged as a target defect area set. Based on the target defect area set and the multi-dimensional matching results, a forming process correction scheme is output. The forming process correction scheme is a forming scheme that requires secondary adjustment of forming parameters during the low-risk forming combination forming process of the irregularly shaped sheet metal parts. This application identifies multiple low-risk forming combinations based on billet basic data, process parameter data, and equipment status data. It then identifies suspected defect areas and, based on multi-dimensional matching results, determines suspected associated influence areas, thus obtaining a set of target defect areas. In identifying the target defect area set within low-risk forming combinations, it not only filters out suspected defect areas but also identifies related suspected influence areas based on their causes. This yields a specific set of target defect areas and associated multi-dimensional matching results, enabling the generation of targeted forming process correction schemes. This method addresses the problem that traditional secondary process corrections are often "one-size-fits-all" adjustments that fail to consider the specific causes of defect areas, resulting in low correction efficiency and a high risk of introducing new defects.

[0007] Secondly, embodiments of this application provide a progressive multi-point forming process correction system for irregularly shaped sheet metal parts, applied to electronic devices, for implementing the progressive multi-point forming process correction method for irregularly shaped sheet metal parts as described in any of the first aspects above. The progressive multi-point forming process correction system for irregularly shaped sheet metal parts includes: The acquisition unit is used to acquire the forming parameters of the irregular sheet metal part; wherein, the forming parameters include basic data of the blank, process parameter data and equipment status data; The combination unit is used to determine multiple low-risk forming combinations based on the billet basic data, the process parameter data, and the equipment status data; A defect unit is used to identify suspected defect areas of irregularly shaped sheet metal parts based on a variety of the aforementioned low-risk forming combinations. The association unit is used to obtain multi-dimensional matching results based on the suspected defect area of ​​the irregular sheet metal part, and determine the suspected associated influence area; wherein, the suspected associated influence area refers to other forming areas that are affected by the defect in the suspected defect area; the multi-dimensional matching result refers to the result obtained by matching the parameter similarity between the local area of ​​the irregular sheet metal part and the suspected defect area from three dimensions: basic data of the blank, process parameter data, and equipment status data; The target unit is used to merge the suspected defect area and the suspected associated influence area of ​​the irregular sheet metal part into a target defect area set; The forming unit is used to output a forming process correction scheme based on the target defect area set and the multi-dimensional matching result; wherein the forming process correction scheme is a forming scheme that requires secondary adjustment of forming parameters during the low-risk forming combination forming process of irregular sheet metal parts.

[0008] Thirdly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method as described in any of the foregoing aspects.

[0009] Fourthly, embodiments of this application provide a computer program product that, when run on an electronic device, causes the electronic device to perform the method described in any one of the first aspects above.

[0010] It is understood that the beneficial effects of the second to fourth aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a flowchart illustrating a progressive multi-point forming process correction method for irregularly shaped sheet metal parts provided in an embodiment of this application. Figure 2This is a schematic diagram of the operation of a progressive multi-point forming process correction method for irregular sheet metal parts provided in an embodiment of this application; Figure 3 This is a schematic diagram of the progressive multi-point forming process correction system for irregularly shaped sheet metal parts provided in the embodiments of this application; Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0013] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0014] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0015] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0016] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0017] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0018] References to "one embodiment" or "some embodiments" as described in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0019] In related technologies, the traditional process of progressive multi-point forming of irregular sheet metal parts often involves a "one-size-fits-all" approach to secondary corrections, failing to develop differentiated solutions based on the specific causes of defects in the affected areas. This results in low correction efficiency and a high risk of introducing new defects. Because the specific causes of defects are not analyzed (whether it's uneven material properties, unreasonable local process parameters, or localized equipment precision deviations), global parameter adjustments can be ineffective. For example, reducing the overall forming speed to address localized wrinkling can decrease forming efficiency in other areas of the part, and may even lead to increased springback due to excessively low deformation rates. Conversely, reducing the overall punch pitch to address localized cracking can significantly increase forming time and reduce production efficiency.

[0020] To address the aforementioned issues, embodiments of this application provide a method for correcting the progressive multi-point forming process of irregularly shaped sheet metal parts. This method includes: acquiring forming parameters of an irregularly shaped sheet metal part; wherein the forming parameters include basic data of the blank, process parameter data, and equipment status data; determining multiple low-risk forming combinations based on the basic data of the blank, process parameter data, and equipment status data; determining suspected defect areas of the irregularly shaped sheet metal part based on the multiple low-risk forming combinations; obtaining multi-dimensional matching results based on the suspected defect areas of the irregularly shaped sheet metal part to determine suspected associated influence areas; wherein the suspected associated influence areas represent other forming areas affected by the defect in the suspected defect area; the multi-dimensional matching results refer to the results obtained by matching the parameter similarity between local areas of the irregularly shaped sheet metal part and suspected defect areas from three dimensions: basic data of the blank, process parameter data, and equipment status data; merging the suspected defect areas and suspected associated influence areas of the irregularly shaped sheet metal part as a target defect area set; and outputting a forming process correction scheme based on the target defect area set and the multi-dimensional matching results; wherein the forming process correction scheme is a forming scheme that requires secondary adjustment of forming parameters during the forming process of the low-risk forming combination of the irregularly shaped sheet metal part. This application identifies multiple low-risk forming combinations based on billet basic data, process parameter data, and equipment status data. It then identifies suspected defect areas and, based on multi-dimensional matching results, determines suspected associated influence areas, thus obtaining a set of target defect areas. In identifying the target defect area set within low-risk forming combinations, it not only filters out suspected defect areas but also identifies related suspected influence areas based on their causes. This yields a specific set of target defect areas and associated multi-dimensional matching results, enabling the generation of targeted forming process correction schemes. This method addresses the problem that traditional secondary process corrections are often "one-size-fits-all" adjustments that fail to consider the specific causes of defect areas, resulting in low correction efficiency and a high risk of introducing new defects.

[0021] The progressive multi-point forming process correction method for irregular sheet metal parts provided in this application embodiment can be applied to electronic devices. In this case, the electronic device is the executing subject of the progressive multi-point forming process correction method for irregular sheet metal parts provided in this application embodiment. This application embodiment does not impose any restrictions on the specific type of electronic device.

[0022] For example, electronic devices can be industrial control computers, edge computing gateways, production scheduling servers, cloud servers, industrial tablets, or intelligent scheduling terminals. Electronic devices include memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the methods described in any of the foregoing aspects.

[0023] To better understand the progressive multi-point forming process correction method for irregular sheet metal parts provided in this application embodiment, the specific implementation process of the progressive multi-point forming process correction method for irregular sheet metal parts provided in this application embodiment will be described by way of example below.

[0024] Figure 1 A flowchart illustrating the progressive multi-point forming process correction method for irregularly shaped sheet metal parts provided in this application embodiment is shown. Figure 2 This illustration shows a schematic diagram of the operation of the progressive multi-point forming process correction method for irregularly shaped sheet metal parts provided in an embodiment of this application. The progressive multi-point forming process correction method for irregularly shaped sheet metal parts includes: S100: Obtain the forming parameters of the irregularly shaped sheet metal part. These forming parameters include basic blank data, process parameter data, and equipment status data.

[0025] It can be understood that forming parameters refer to the complete set of core data affecting the progressive multi-point forming quality of irregular sheet metal parts, and are the basis for subsequent screening of low-risk forming combinations. Specifically: Basic billet data refers to the inherent property data of the sheet metal billet, which can be obtained through material testing instruments. For example, the thickness uniformity of 6061 aluminum alloy billet measured by a universal testing machine is 2.0±0.05mm, which can be measured by a laser thickness gauge. Process parameter data refers to the adjustable operating parameters during the forming process, which can be preset using forming simulation software based on the surface requirements of the irregular sheet metal part (such as a double-curvature irregular part), such as a punch pitch of 0.5mm and a forming speed of 5mm / s. Equipment status data refers to the operating performance data of the forming equipment, which can be collected in real time through the equipment's built-in sensors, such as a punch positioning accuracy deviation ≤±0.02mm.

[0026] S200 determines a variety of low-risk forming combinations based on billet basic data, process parameter data, and equipment status data.

[0027] Understandably, low-risk forming combinations refer to parameter combinations where the matching degree of billet, process, and equipment parameters is high, and the probability of defects such as wrinkling and cracking during forming is less than 10%. This can be obtained through feature extraction combined with model judgment. For example, based on the above 6061 aluminum alloy billet data, three parameter combinations were selected: billet thickness uniformity of 2.0±0.05mm, punch pitch of 0.4-0.6mm, forming speed of 4-6mm / s, and punch positioning accuracy deviation ≤±0.02mm. All of these belong to low-risk forming combinations.

[0028] As an optional embodiment of this application, in step S200, based on the billet basic data, process parameter data, and equipment status data, several low-risk forming combinations are determined, including: S210 extracts the first forming quality feature from the basic data of the billet, the process parameter data, and the equipment status data.

[0029] It can be understood that the first forming quality feature refers to a quantitative feature vector that comprehensively reflects the matching rationality of "bill basic data, process parameter data, and equipment status data". It can be obtained by first preprocessing the data, then extracting it by dimension, and then weighted and fusing it. For example, based on the above 6061 aluminum alloy billet data, preset process parameters, and equipment status data, the extracted first forming quality feature is [0.92, 0.88, 0.95] (corresponding to the matching quantification values ​​of the billet, process, and equipment dimensions, respectively).

[0030] In one possible implementation, S210, the first forming quality feature is extracted from the billet basic data, process parameter data, and equipment status data, including: S211, standardize and preprocess the basic data of the billet, process parameter data, and equipment status data to obtain standard forming parameters.

[0031] Standardization preprocessing can be understood as the process of eliminating differences in the dimensions of data across different dimensions and unifying the data range. This can be achieved by first standardizing units and then verifying their reasonableness. Standard forming parameters refer to parameters that, after preprocessing, conform to industry standards or historical best-fit ranges, thereby improving the accuracy of standard forming parameters.

[0032] For example, in S211, the basic data of the billet, the process parameter data, and the equipment status data are respectively subjected to standardized preprocessing to obtain standard forming parameters, including: S2111 performs unit consistency processing on the billet basic data, process parameter data, and equipment status data respectively to obtain multiple basic unit data.

[0033] Unit consistency processing is understandable; it refers to the process of converting similar data with different units into a unified unit, which can be achieved through unit conversion formulas. The basic unit data is the original data after unit unification, avoiding subsequent analytical biases caused by unit differences. For example, unifying the blank thickness ("mm"), process speed ("mm / s"), and equipment positioning accuracy ("μm") to "mm" yields the following basic unit data: blank thickness 2.0mm, forming speed 5mm / s, punch positioning accuracy 0.02mm.

[0034] S2112 predicts rationality characteristics from multiple basic unit data and acquired unit data templates. The unit data template refers to a pre-established dataset of optimal matching relationships between billet basic data, process parameter data, and equipment status data, based on historical forming data or industry standards. Rationality characteristics reflect the degree of similarity between the basic unit data and the acquired unit data templates.

[0035] It is understandable that the unit data template can be constructed by compiling successful forming data (a total of 1000 sets) of similar 6061 aluminum alloy irregular parts from the past 3 years. For example, the optimal value for the 6061 aluminum alloy billet thickness in the unit data template is 2mm, the optimal forming speed is 5mm / s, and the optimal punch positioning accuracy is 0.02mm. The rationality feature is a quantified value (range 0-1) obtained by calculating the overlap between the basic unit data and the optimal value of the unit data template. It can be calculated using the cosine similarity algorithm. For example, if the overlap between the basic unit data (billet thickness 2.0mm, forming speed 5mm / s, punch positioning accuracy 0.02mm) and the template is 1, then the rationality feature is 1.

[0036] S2113, based on the rationality characteristics, the standard forming parameters are obtained.

[0037] It is understandable that by setting a reasonable feature threshold (such as a threshold ≥ 0.8) to filter basic unit data, the basic unit data that meets the threshold requirement are the standard forming parameters. For example, if reasonable feature 1 ≥ 0.8, the corresponding standard forming parameters are: blank thickness 2.0mm, forming speed 5mm / s, and punch positioning accuracy 0.02mm.

[0038] It should be noted that the rationality feature threshold was determined based on statistical analysis of historical forming data and verification of process failure boundaries. A correlation model between rationality features and forming pass rate was constructed using 1000 sets of historical data. When the target pass rate is ≥95%, the corresponding minimum rationality feature value is 0.8. After verification through 20 sets of repeatable tests, the forming pass rate with the rationality feature threshold is consistently ≥96.2%, therefore, the rationality feature threshold is determined to be 0.8.

[0039] By employing the steps S2111 to S2113 described above, it is helpful to eliminate differences in data units and interference from abnormal data, making the extracted first forming quality features more accurate and providing a reliable data foundation for the screening of low-risk forming combinations.

[0040] S212, based on standard forming parameters, extracts the characteristics of billet performance, process parameters, and equipment status.

[0041] It is understandable that the billet performance dimension characteristic is a quantitative indicator reflecting the billet's suitability for forming requirements. This can be obtained through statistical analysis of billet data in standard forming parameters. For example, based on standard 6061 aluminum alloy billet data, the extracted billet performance dimension characteristic is [thickness uniformity 0.05mm (0.93 after standardization)]. The process parameter dimension characteristic is a quantitative indicator reflecting the rationality of process parameters. This can be calculated by the matching degree between process parameters and billet characteristics. For example, the extracted process parameter dimension characteristic is [forming speed 5mm / s (0.92 after standardization)]. The equipment status dimension characteristic is a quantitative indicator reflecting equipment stability. This can be calculated by the adaptability between equipment parameters and process requirements. For example, the extracted equipment status dimension characteristic is [positioning accuracy 0.02mm (0.96 after standardization)].

[0042] S213, weighted fusion of the billet performance dimension feature, process parameter dimension feature and equipment status dimension feature to obtain the first forming quality feature.

[0043] As can be understood, weighted fusion refers to calculating the weighted sum of the characteristics of each dimension based on their influence on forming quality (determined through the analytic hierarchy process, for example, a weight of 0.4 for billet performance, 0.3 for process parameters, and 0.3 for equipment status). The resulting comprehensive quantitative value is the first forming quality characteristic. For example, based on the above characteristics, the first forming quality characteristic is calculated as: 0.93 × 0.4 + 0.92 × 0.3 + 0.96 × 0.3 = 0.936.

[0044] By employing the steps S211 to S213 described above, it is helpful to transform scattered multi-dimensional data into comprehensive features that can be directly used for model determination, thereby improving the efficiency and accuracy of screening low-risk formed combinations.

[0045] S220, input the first forming quality feature into a preset first forming quality classification model to obtain the low-risk forming probability of the irregular sheet metal part. The first forming quality classification model is a learning model that takes the first forming quality feature as input and the low-risk forming probability corresponding to the first forming quality feature as the expected output.

[0046] It is understandable that the first forming quality classification model is a binary classification model built on the random forest algorithm, trained using historical forming data (the training samples consist of 1000 sets of "first forming quality feature - low-risk forming probability" data). The low-risk forming probability is the probability that the current forming combination belongs to low risk. For example, inputting the first forming quality feature of 0.936 into the model yields a low-risk forming probability of 0.95.

[0047] It should be noted that the first forming quality classification model is a binary classification model. Its core objective is to input the first forming quality features and output the low-risk forming probability of the forming combination, thus determining whether the parameter combination is a low-risk forming combination. For example, the complete training process for the forming scenario of 6061 aluminum alloy irregular parts first collects historical data on the progressive multi-point forming of similar 6061 aluminum alloy irregular parts from the past three years, along with orthogonal experimental data, totaling 1000 valid samples. The Random Forest (RF) algorithm is preferentially chosen because it has strong fitting ability for nonlinear features and can effectively avoid overfitting, making it suitable for risk determination scenarios involving forming parameter combinations. The training set is input into the initialized Random Forest model, and training is performed tree by tree. Each decision tree selects the optimal splitting feature based on the Gini coefficient to generate decision rules. After all decision trees are trained, the low-risk forming probability of the samples is output through a voting mechanism. The trained model is embedded into the parameter filtering module of the forming equipment, receiving the first forming quality features in real time and outputting the low-risk forming probability. For every 200 new forming data sets, incremental training is performed on the model, updating the feature weights of the decision trees to adapt to new forming materials or processes.

[0048] S230, the combination of forming parameters whose low-risk forming probability exceeds the set first classification threshold is determined as a low-risk forming combination.

[0049] It is understandable that the first classification threshold is a judgment threshold set based on historical forming pass rate requirements (determined through experiments, for example, a threshold set to 0.85). When the low-risk forming probability exceeds the first classification threshold, it indicates that the combination of forming parameters has a low risk of forming defects and can be judged as a low-risk forming combination. For example, a low-risk forming probability of 0.95 > 0.85 corresponds to the combination of "2.0mm billet thickness, 0.5mm punch pitch, 5mm / s forming speed, and 0.02mm punch positioning accuracy" as a low-risk forming combination. Forming parameter combinations with a low-risk forming probability of less than or equal to 0.85 are judged as high-risk combinations, and optimization prompts are output for billet basic data, process parameter data, or equipment status data. For example, if the billet thickness uniformity does not meet the standard, it is recommended to replace the billet.

[0050] By employing the steps S210 to S230 described above, it is helpful to accurately screen out low-risk combinations from a large number of forming parameter combinations, thereby reducing the probability of forming defects from the source.

[0051] S300 identifies potential defect areas in irregularly shaped sheet metal parts based on a variety of low-risk forming combinations.

[0052] It is understandable that a suspected defect area refers to a localized area of ​​a sheet metal part that shows signs of defect during the forming process (such as signs of impending wrinkling or localized stress concentration). For example, when forming based on the aforementioned low-risk forming combination, if abnormal stress values ​​are detected in a localized area of ​​the sheet metal part through monitoring, that localized area is determined to be a suspected defect area.

[0053] As an optional embodiment of this application, S300, based on multiple low-risk forming combinations, identifies suspected defect areas in irregularly shaped sheet metal parts, including: S310, acquire forming process data of low-risk forming assembly, including local stress values ​​of sheet metal parts, surface forming deviations, temperature distribution data and surface morphology image data.

[0054] It can be understood that forming process data refers to dynamic data reflecting the state and forming effect of the sheet metal part during the forming process. This data can be collected under low-risk forming conditions: blank thickness 2.0mm, forming speed 5mm / s, and punch positioning accuracy 0.02mm. Specifically, local stress values ​​can be obtained by measuring the contact pressure between the punch and the sheet metal part using a miniature pressure sensor embedded in the top of the punch in the forming equipment. This measurement, combined with the contact area and Hertzian contact theory, allows for reverse calculation using Hertzian contact theory formulas. Reverse the surface stress of sheet metal parts ( Here, F represents the local stress value, r is the contact force measured by the sensor, t is the punch radius, and E is the sheet metal thickness. For example, the punch radius... =2mm, sheet metal thickness =2.0mm, the sensor measured the contact force =100N, elastic modulus of billet =69GPa, the calculated local stress value is... Surface forming deviation can be obtained using a laser displacement sensor (measurement accuracy ±0.01mm), for example, a local forming deviation of 0.15mm can be measured. Temperature distribution data can be obtained using an infrared thermal imager (temperature range -20~500℃), for example, a local temperature of 120℃ can be measured. Surface morphology image data can be obtained using a high-speed camera to obtain surface texture images of local areas, and then the surface texture entropy (1.8) can be obtained.

[0055] S320 performs in-depth analysis of local stress values, surface forming deviations, temperature distribution data, and surface morphology image data of sheet metal parts to obtain the second forming quality characteristics.

[0056] It can be understood that the second forming quality feature refers to a quantitative feature vector that can comprehensively reflect the defect risk of a local area of ​​a sheet metal part. For example, based on the in-depth analysis of the local stress value, surface forming deviation, temperature distribution data and surface morphology image data of the sheet metal part, the extracted second forming quality feature is [0.82].

[0057] In one possible implementation, S320 involves in-depth analysis of local stress values, surface forming deviations, temperature distribution data, and surface morphology image data of the sheet metal part to obtain a second forming quality characteristic, including: S321 involves outlier removal and smoothing filtering of local stress values, surface forming deviations, and temperature distribution data for sheet metal parts to obtain stable numerical data. Based on this stable numerical data, numerical defect features are extracted. These numerical defect features are quantifiable parameters reflecting the degree of defect generation and development.

[0058] It is understandable that outlier removal can use the 3σ criterion (removing data that exceeds the mean ± 3 times the standard deviation, such as removing outlier data of 450MPa in stress values), and smoothing filtering can use moving average filtering. The stabilized numerical data obtained after processing is a single specific value: stabilized stress value 245MPa, stabilized forming deviation 0.109mm, and stabilized temperature 95℃. Numerical defect characteristics of local areas of sheet metal parts can be obtained by calculating the local stress value, surface forming deviation, and temperature distribution data of the sheet metal part, and dividing them by the corresponding stable numerical data to obtain stress matching coefficients of 1.204, forming deviation matching coefficients of 1.376, and temperature matching coefficients of 1.263. Based on the statistical range of each coefficient from historical data (stress [0.8, 1.3], forming deviation [0.7, 1.5], temperature [0.8, 1.28]), the matching coefficients are mapped to the [0, 1] interval through a standardization formula, resulting in standardized coefficients of 0.808, 0.845, and 0.965, respectively. Then, the numerical defect characteristics are calculated using preset weights (stress 0.4, forming deviation 0.4, temperature 0.2). For example, numerical defect characteristic = 0.808 × 0.4 + 0.845 × 0.4 + 0.965 × 0.2 ≈ 0.323 + 0.338 + 0.193 = 0.854 ≈ 0.85.

[0059] S322: The surface morphology image data is processed by grayscale conversion, denoising, and image enhancement to obtain sharpened image data. Based on the sharpened image data, image-type defect features are extracted. Among them, image-type defect features are image quantization parameters that characterize the morphology and distribution of surface defects.

[0060] It is understandable that grayscale conversion transforms a color image into a grayscale image. Denoising can be achieved using median filtering (3×3 window size), and image enhancement can be achieved using histogram equalization. The resulting clearer image data can clearly present surface texture. Image-based defect features can be obtained through image texture analysis, such as extracting the surface texture entropy (1.8) and the historical range of the surface texture entropy (1.0-2.03). Image-based defect features can be derived from a standardized formula. We obtained, among which, It is an image-type defect feature. This represents the maximum value of the historical range of surface texture entropy. This represents the minimum historical range of surface texture entropy. The surface texture entropy extracted was normalized to obtain an image-type defect feature of 0.78.

[0061] S323: After spatial alignment according to the spatial coordinates of the forming area of ​​the irregular sheet metal part, the numerical defect features and the image defect features are weighted and fused to obtain the second forming quality feature.

[0062] It can be understood that the forming area of ​​an irregularly shaped sheet metal part refers to several independent regions into which the sheet metal part is divided according to a 5mm×5mm grid rule (for example, a 100mm×100mm sheet metal part is divided into 400 forming regions), and each region corresponds to a unique spatial coordinate (e.g., (x1,y1)=(5,5)). Spatial alignment refers to matching numerical features and image features with the same spatial coordinates. Weighted fusion is calculated by summing the weights of the two types of features based on their defect identification weights (numerical weight 0.6, image weight 0.4) to obtain the second forming quality feature, for example, 0.85×0.6+0.78×0.4=0.82.

[0063] By employing the steps S321 to S323 described above, it is helpful to integrate multi-source forming process data, extract comprehensive features that can accurately reflect defect risks, and provide a reliable basis for the determination of suspected defect areas.

[0064] S330, input the second forming quality feature into the second forming quality classification model to obtain the defect probability of the forming area of ​​the irregular sheet metal part. Here, the forming area of ​​the irregular sheet metal part represents the area allocated to the irregular sheet metal part according to preset rules. The second forming quality classification model is a machine learning model that takes the second forming quality feature as input and the defect probability corresponding to the second forming quality feature as the expected output.

[0065] It is understandable that the second forming quality classification model is a multi-classification model built based on a CNN+LSTM hybrid algorithm, and is trained using second forming quality features and defect probabilities (the training samples consist of 800 sets of "second forming quality features - defect probabilities" data). The defect probability is the confidence level that a defect exists in the current forming area (the value ranges from 0 to 1). For example, inputting the aforementioned second forming quality feature of 0.82 into the model yields a defect probability of 0.76.

[0066] It should be noted that the second forming quality classification model is a probabilistic output model. Its core objective is to input second forming quality features and output the defect probability of the forming area to determine whether it is a suspected defect area. For example, considering the complete training process of forming 6061 aluminum alloy irregular parts, firstly, second forming quality feature data and defect probability data during the forming process are collected, totaling 800 sets of valid samples. A CNN+LSTM hybrid algorithm is selected to fully integrate the advantages of multi-source data and improve defect recognition accuracy. The training set is input into the hybrid model; the CNN branch extracts image features, and the LSTM branch extracts temporal features. The fusion layer concatenates the two types of features and maps them to the defect type space through a fully connected layer. The cross-entropy loss function is used to calculate the error between the predicted value and the true label, and the loss value is minimized using the Adam optimizer with a learning rate of 0.001. Iterative training is performed for 100 epochs, with model accuracy validated every 10 epochs. If the validation set loss value does not decrease for 20 consecutive epochs, training is stopped. The trained model is embedded into the edge computing module of the forming equipment to receive second forming quality features in real time and output the defect probability.

[0067] S340, the forming area of ​​irregular sheet metal parts with a defect probability exceeding the set second classification threshold is identified as a suspicious defect area.

[0068] It is understandable that the second classification threshold is a judgment threshold set based on the defect recognition accuracy requirements (determined through experiments, for example, a threshold of 0.63). When the defect probability exceeds the second classification threshold, it indicates that there are obvious signs of defects in the forming area of ​​the irregular sheet metal part, and it can be judged as a suspected defect area. For example, if the defect probability is 0.76 > 0.63, the forming area with spatial coordinates (5,5) is the suspected defect area.

[0069] It should be noted that the second classification threshold needs to be differentiated because different defect types (such as cracking, wrinkling, and step effect) have different degrees of impact on part performance. The higher the severity, the lower the threshold. The threshold must match the recognition accuracy of the second forming quality classification model to avoid model failure due to excessively high or low thresholds. Based on the test set validation results of the second forming quality classification model, the precision and recall rates under different defect probability thresholds are statistically analyzed: Recall rate: Number of defect areas correctly identified by the model ÷ Total number of actual defect areas. Precision rate: Number of defect areas correctly identified by the model ÷ Total number of defect areas judged by the model. The validation results meet the production quality requirements, and the differentiated second classification thresholds are determined as follows: Crack defect threshold is 0.50, wrinkling defect threshold is 0.65, step effect threshold is 0.75, and the second classification threshold of 0.63 is a simplified value for a general scenario obtained by taking the average of the differentiated thresholds. This simplified value for a general scenario is suitable for batch judgment of defects with unclear types.

[0070] By employing the steps S310 to S340 described above, it is possible to locate suspected defect areas in real time and accurately during the forming process, thereby preventing the defects from expanding further.

[0071] S400: Based on the suspected defect areas of the irregularly shaped sheet metal parts, multi-dimensional matching results are obtained to determine the suspected associated influence areas. The suspected associated influence areas refer to other formed areas affected by the defect in the suspected defect area. The multi-dimensional matching results refer to the results obtained by matching the parameter similarity between local areas of the irregularly shaped sheet metal parts and the suspected defect areas from three dimensions: basic blank data, process parameter data, and equipment status data.

[0072] It is understandable that a suspected associated influence area refers to an area that is highly similar to a suspected defect area in terms of raw material, process, and equipment parameters, and although it currently has no defects, it has the potential risk of defects. The multi-dimensional matching result is a comprehensive quantitative value of the similarity of parameters in each dimension (range 0-1).

[0073] As an optional embodiment of this application, in step S400, based on the suspected defect areas of the irregularly shaped sheet metal parts, multi-dimensional matching results are obtained to determine the suspected associated influence areas, including: S410 determines multi-dimensional matching rules based on the suspected defect areas of irregularly shaped sheet metal parts.

[0074] As we can understand it, multi-dimensional matching rules refer to setting the similarity calculation methods and weights for each dimension parameter. For example, the basic data of billet is calculated using cosine similarity (weight 0.3), the process parameter data is calculated using Euclidean distance similarity (weight 0.4), and the equipment status data is calculated using Pearson correlation coefficient (weight 0.3). The comprehensive matching result = billet similarity × 0.3 + process similarity × 0.4 + equipment similarity × 0.3. Multi-dimensional matching rules can be verified and optimized through historical defect association data to improve the accuracy of matching results.

[0075] It should be noted that Euclidean distance, with larger numerical values, indicates greater parameter differences, while similarity, with larger numerical values, indicates greater parameter similarity. When using Euclidean distance, it is necessary to convert it to a similarity score within the 0-1 range using a conversion formula before incorporating it into multi-dimensional matching rules. The conversion formula is as follows: ,in The similarity is determined by Euclidean distance. For Euclidean distance, This is the maximum Euclidean distance in the historical data, that is, the distance value with the greatest difference among all process parameter data.

[0076] S420 matches the multi-dimensional matching rules to the forming area of ​​the irregular sheet metal part to obtain the multi-dimensional matching results of the suspected defect area.

[0077] It can be understood that by substituting the blank, process, and equipment parameters of each forming region with the corresponding parameters of the suspected defect region into the matching rules, the comprehensive matching result of each forming region can be calculated. For example, for the forming region at spatial coordinates (10,10), the similarity between the blank parameters (6061 aluminum alloy, blank thickness 2.0mm) of the forming region and the blank parameters of the suspected defect region is 0.98, the similarity between the process parameters (punch pitch 0.5mm, forming speed 5.2mm / s) is 0.92, and the similarity between the equipment parameters (punch positioning accuracy 0.02mm) is 0.96. Substituting these into the rules, the multi-dimensional matching result of the suspected defect region is calculated as 0.98×0.3+0.92×0.4+0.96×0.3=0.95.

[0078] S430 identifies irregularly shaped sheet metal parts whose multi-dimensional matching results exceed a set matching threshold as suspicious correlation influence areas.

[0079] It is understandable that the matching degree threshold is a threshold set based on the needs of potential defect risk management (determined through experiments, for example, a threshold of 0.85). When the multi-dimensional matching result exceeds the matching degree threshold, it indicates that the parameters of the formed region and the suspected defect region are highly similar, indicating a potential defect risk, and can be identified as a suspected associated influence region. For example, if the matching result 0.95 > 0.85, the formed region corresponding to the spatial coordinates (10, 10) is the suspected associated influence region.

[0080] By employing the steps S410 to S430 described above, it is possible to comprehensively identify potential risk areas related to suspected defect areas and achieve full-range control over defect risks.

[0081] S500 merges the suspected defect areas and suspected associated influence areas of irregularly shaped sheet metal parts into a set of target defect areas.

[0082] It can be understood that the target defect area set refers to the set of all areas that need to be modified in the process. For example, by merging the suspected defect area (5,5) and the suspected associated influence areas (10,10) and (15,15) mentioned above, we get the target defect area set {(5,5), (10,10), (15,15)}.

[0083] As an optional embodiment of this application, in S500, the suspected defect areas and suspected associated influence areas of the irregularly shaped sheet metal parts are merged into a target defect area set, including: S510: Obtain the spatial coordinates, defect type, and defect severity information of the suspected defect area, and establish a suspected defect area information set.

[0084] It is understandable that the defect type can be obtained through the output of the second forming quality classification model (such as "wrinkling"), and the defect severity can be quantified by the defect probability (such as a defect probability of 0.76 corresponding to "minor defect"). The suspected defect area information set is a dataset containing the core information of each suspected defect area, for example, the set is {(5,5): process parameter matching, wrinkling, minor}.

[0085] S520: Obtain the spatial coordinates of the suspected associated area and the matching dimension information with the suspected defect area, and establish a set of information on the suspected associated area.

[0086] It can be understood that matching dimension information refers to parameter dimensions (such as "process parameter dimension") that are highly similar to the suspected defect area. The suspected associated influence area information set is a dataset containing the core information of each suspected associated influence area, for example, the set is {(10,10): process parameter matching, potential wrinkling, potential. (15,15): equipment parameter matching, potential wrinkling, potential}.

[0087] S530, the set of information on suspected defective areas and the set of information on suspected related areas are deduplicated and integrated to form the set of target defective areas.

[0088] Deduplication and integration can be understood as removing duplicate regions from two sets (if regions share the same spatial coordinates), and summarizing the information of the remaining regions to form a target defect region set containing region coordinates, defect status, and risk causes. For example, the integrated target defect region set might be {(5,5): process parameter matching, wrinkling, mild; (10,10): process parameter matching, potential wrinkling, potential; (15,15): equipment parameter matching, potential wrinkling, potential}.

[0089] By adopting the above steps S510 to S530, it is helpful to systematically sort out the areas and core information that need to be corrected, and provide a clear basis for the subsequent formulation of precise process correction methods.

[0090] S600, based on the target defect region set and multi-dimensional matching results, outputs a forming process correction scheme. This scheme addresses the low-risk forming assembly process of irregularly shaped sheet metal parts that requires secondary adjustments to forming parameters.

[0091] It can be understood that the forming process correction scheme refers to a specific scheme for secondary adjustment of the parameters of the original low-risk forming combination based on the defect cause of the target defect area (determined based on multi-dimensional matching results). For example, for the above target defect area, the output correction method is: "(5,5) area: adjust the punch pitch to 0.4mm. (10,10) area: adjust the forming speed to 4.5mm / s. (15,15) area: compensate the punch positioning accuracy to 0.01mm."

[0092] As an optional embodiment of this application, S600, based on the target defect region set and multi-dimensional matching results, outputs a forming process correction scheme, including: S610, analyze the attribute information of the forming areas of each irregular sheet metal part in the target defect area set. The attribute information includes spatial coordinates, defect type, and defect severity information.

[0093] It can be understood that parsing attribute information refers to extracting the core attributes of each region from the target defect region set to clarify the defect status of each region. For example, parsing the attribute information of (5,5) yields: coordinates (5,5), defect type: wrinkling, severity: mild. Parsing the attribute information of (10,10) yields: coordinates (10,10), defect type: potential wrinkling, severity: potential.

[0094] S620, based on spatial coordinates, defect type and defect severity information combined with multi-dimensional matching results, locates the cause of defects corresponding to each target defect region in the target defect region set.

[0095] It is understandable that the cause of a defect refers to the core factors that lead to a defect or potential defect, which can be derived by reverse inference from the results of multi-dimensional matching. For example, in the multi-dimensional matching results of region (5,5), the process parameter similarity is the highest (0.92). Combined with the defect type wrinkling, the cause of the defect is located as "excessive punch pitch (0.5mm) leading to excessive local deformation". In the multi-dimensional matching results of region (15,15), the equipment parameter similarity is the highest (0.96). The cause of the defect is located as "punch positioning accuracy deviation (0.02mm) leading to forming offset".

[0096] S630 generates differentiated process correction strategies based on different defect causes.

[0097] It is understandable that differentiated process correction strategies refer to targeted parameter adjustment schemes formulated based on different defect causes, with the parameter adjustment range determined based on historical correction data and simulation verification. For example, for the cause of "excessive punch pitch," a correction strategy is generated to "adjust the punch pitch from 0.5mm to 0.4mm." For the cause of "excessive forming speed," a correction strategy is generated to "adjust the forming speed from 5mm / s to 4.5mm / s." For the cause of "punch positioning accuracy deviation," a correction strategy is generated to "compensate for punch positioning accuracy, adjusting it from 0.02mm to 0.01mm."

[0098] S640 associates the process correction strategy with the corresponding target defect area to generate a forming process correction scheme.

[0099] This can be understood as binding the spatial coordinates of each target defect area with the corresponding process correction strategy, forming a specific method that can directly guide the adjustment of the forming equipment. For example, the generated forming process correction scheme is as follows: 1. Coordinate (5,5) area: Adjust the punch pitch to 0.4mm. 2. Coordinate (10,10) area: Adjust the forming speed to 4.5mm / s. 3. Coordinate (15,15) area: Compensate the punch positioning accuracy to 0.01mm.

[0100] By adopting the above steps S610 to S640, it is helpful to improve the efficiency of defect correction and forming quality, while avoiding new defects caused by "one-size-fits-all" adjustments.

[0101] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0102] Corresponding to the progressive multi-point forming process correction method for irregular sheet metal parts described in the above embodiments, this application also provides a progressive multi-point forming process correction system for irregular sheet metal parts. Each unit of the system can realize each step of the progressive multi-point forming process correction method for irregular sheet metal parts. Figure 3 The diagram shows a structural block diagram of the progressive multi-point forming process correction system for irregular sheet metal parts provided in the embodiments of this application. For ease of explanation, only the parts related to the embodiments of this application are shown.

[0103] Reference Figure 3 The system includes: The acquisition unit is used to acquire the forming parameters of irregularly shaped sheet metal parts. These forming parameters include basic blank data, process parameter data, and equipment status data.

[0104] The combination unit is used to determine a variety of low-risk forming combinations based on billet basic data, process parameter data, and equipment status data.

[0105] Defect cells are used to identify potential defect areas in irregularly shaped sheet metal parts based on a variety of low-risk forming combinations.

[0106] The association unit is used to obtain multi-dimensional matching results based on the suspected defect areas of irregularly shaped sheet metal parts, and to determine the suspected associated influence areas. The suspected associated influence areas refer to other formed areas that are affected by the defect in the suspected defect area. The multi-dimensional matching results refer to the results obtained by matching the parameter similarity between local areas of the irregularly shaped sheet metal part and the suspected defect areas from three dimensions: basic blank data, process parameter data, and equipment status data.

[0107] The target unit is used to merge the suspected defect areas and suspected associated influence areas of irregular sheet metal parts into a target defect area set.

[0108] The forming unit is used to output a forming process correction scheme based on the target defect region set and multi-dimensional matching results. This forming process correction scheme is a forming scheme that requires secondary adjustment of forming parameters during the low-risk forming assembly process of irregularly shaped sheet metal parts.

[0109] It should be noted that the information interaction and execution process between the above-mentioned units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, which will not be repeated here.

[0110] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units is merely an example. In practical applications, the above functions can be assigned to different functional units as needed, that is, the internal structure of the system can be divided into different functional units to complete all or part of the functions described above. The functional units in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0111] This application also provides an electronic device. Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 4 As shown, the electronic device 6 of this embodiment includes: at least one processor 60 ( Figure 4 Only one is shown in the image), at least one memory 61 ( Figure 4 (Only one is shown in the image) and a computer program 62 stored in the at least one memory 61 and executable on the at least one processor 60, wherein when the processor 60 executes the computer program 62, it causes the electronic device 6 to perform the steps in any of the above embodiments of the progressive multi-point forming process correction method for irregular sheet metal parts, or causes the electronic device 6 to perform the functions of the units in the above embodiments of the systems.

[0112] For example, the computer program 62 may be divided into one or more units, which are stored in the memory 61 and executed by the processor 60 to complete this application. The one or more units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program 62 in the electronic device 6.

[0113] The electronic device 6 may be an industrial control computer, an edge computing gateway, a production scheduling server, a cloud server, an industrial tablet computer, or an intelligent scheduling terminal, etc. The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the method described in any of the foregoing aspects. The electronic device 6 may include, but is not limited to, a processor 60 and a memory 61. Those skilled in the art will understand that... Figure 4 This is merely an example of electronic device 6 and does not constitute a limitation on electronic device 6. It may include more or fewer components than shown, or combine certain components, or different components. For example, it may also include input / output devices, network access devices, buses, etc.

[0114] The processor 60 can be a Central Processing Unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0115] In some embodiments, the memory 61 may be an internal storage unit of the electronic device 6, such as a hard disk or memory of the electronic device 6. In other embodiments, the memory 61 may be an external storage device of the electronic device 6, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 6. Furthermore, the memory 61 may include both internal and external storage units of the electronic device 6. The memory 61 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory 61 can also be used to temporarily store data that has been output or will be output.

[0116] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps in any of the above method embodiments.

[0117] This application provides a computer program product that, when run on an electronic device, causes the electronic device to perform the steps in any of the above method embodiments.

[0118] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or system capable of carrying computer program code to an electronic device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0119] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0120] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0121] In the embodiments provided in this application, it should be understood that the disclosed progressive multi-point forming process correction method, system, and electronic device for irregularly shaped sheet metal parts can be implemented in other ways. For example, the embodiments of the progressive multi-point forming process correction system and electronic device for irregularly shaped sheet metal parts described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, and the indirect coupling or communication connection between units may be electrical, mechanical, or other forms.

[0122] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0123] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for correcting the progressive multi-point forming process of irregularly shaped sheet metal parts, characterized in that, include: Obtain the forming parameters of the irregular sheet metal part; wherein, the forming parameters include basic data of the blank, process parameter data and equipment status data; Based on the billet data, the process parameter data, and the equipment status data, several low-risk forming combinations are determined. Based on various low-risk forming combinations, the suspected defect areas of irregularly shaped sheet metal parts are identified; Based on the suspected defect area of ​​the irregular sheet metal part, a multi-dimensional matching result is obtained to determine the suspected associated influence area; wherein, the suspected associated influence area refers to other forming areas that are affected by the defect in the suspected defect area; the multi-dimensional matching result refers to the result obtained by matching the parameter similarity between the local area of ​​the irregular sheet metal part and the suspected defect area from three dimensions: basic data of the blank, process parameter data, and equipment status data. The suspected defect area and the suspected associated influence area of ​​the irregular sheet metal part are merged into a target defect area set; Based on the target defect region set and the multi-dimensional matching results, a forming process correction scheme is output; wherein, the forming process correction scheme is a forming scheme that requires secondary adjustment of forming parameters during the low-risk forming combination forming process of irregular sheet metal parts.

2. The method for correcting the progressive multi-point forming process of irregularly shaped sheet metal parts according to claim 1, characterized in that, Based on the billet data, process parameter data, and equipment status data, several low-risk forming combinations are determined, including: Extract the first forming quality feature from the billet basic data, the process parameter data, and the equipment status data; The first forming quality feature is input into a preset first forming quality classification model to obtain the low-risk forming probability of the irregular sheet metal part; wherein, the first forming quality classification model is a learning model that takes the first forming quality feature as input and the low-risk forming probability corresponding to the first forming quality feature as the expected output. The combination of forming parameters whose low-risk forming probability exceeds a set first classification threshold is determined as a low-risk forming combination.

3. The method for correcting the progressive multi-point forming process of irregularly shaped sheet metal parts according to claim 2, characterized in that, The extraction of the first forming quality feature from the basic data of the billet, the process parameter data, and the equipment status data includes: Standardized preprocessing is performed on the basic data of the billet, the process parameter data, and the equipment status data to obtain standard forming parameters; Based on standard forming parameters, we extract the characteristics of billet performance, process parameters, and equipment status. The first forming quality feature is obtained by weighted fusion of the billet performance dimension feature, the process parameter dimension feature, and the equipment status dimension feature.

4. The method for correcting the progressive multi-point forming process of irregularly shaped sheet metal parts according to claim 3, characterized in that, The standardization preprocessing of the basic data of the billet, the process parameter data, and the equipment status data to obtain standard forming parameters includes: The basic data of the billet, the process parameter data, and the equipment status data are processed to ensure unit consistency, resulting in multiple basic unit data. Reasonableness features are predicted from multiple basic unit data and acquired unit data templates; wherein, the unit data template refers to the optimal matching relationship dataset of billet basic data, process parameter data, and equipment status data pre-established based on historical forming data or industry standards; the reasonableness features reflect the degree of similarity between the basic unit data and the acquired unit data templates; Based on the aforementioned rationality characteristics, standard forming parameters are obtained.

5. The method for correcting the progressive multi-point forming process of irregularly shaped sheet metal parts according to claim 1, characterized in that, The method of identifying suspected defect areas in irregularly shaped sheet metal parts based on multiple low-risk forming combinations includes: Obtain forming process data of the low-risk forming combination, wherein the forming process data includes local stress values ​​of sheet metal parts, surface forming deviations, temperature distribution data, and surface morphology image data. By deeply analyzing the local stress value of the sheet metal part, the surface forming deviation, the temperature distribution data, and the surface morphology image data, a second forming quality characteristic is obtained. The second forming quality feature is input into the second forming quality classification model to obtain the defect probability of the forming area of ​​the irregular sheet metal part; wherein, the forming area of ​​the irregular sheet metal part represents the area of ​​the irregular sheet metal part allocated according to a preset rule; the second forming quality classification model is a machine learning model that takes the second forming quality feature as input and the defect probability corresponding to the second forming quality feature as the expected output. The irregularly shaped sheet metal parts whose defect probability exceeds the set second classification threshold are identified as suspected defect areas.

6. The method for correcting the progressive multi-point forming process of irregularly shaped sheet metal parts according to claim 5, characterized in that, The in-depth analysis of the local stress value of the sheet metal part, the surface forming deviation, the temperature distribution data, and the surface morphology image data yields a second forming quality characteristic, including: Outlier removal and smoothing filtering are performed on the local stress values, surface forming deviations, and temperature distribution data of sheet metal parts to obtain stable numerical data. Based on the stable numerical data, numerical defect features are extracted; wherein, the numerical defect features reflect quantifiable parameters that reflect the degree of defect generation and development. The surface morphology image data is subjected to grayscale conversion, denoising, and image enhancement processing to obtain sharpened image data. Based on the sharpened image data, image-type defect features are extracted; wherein, the image-type defect features are image quantization parameters that characterize the morphology and distribution of surface defects. After spatial alignment according to the spatial coordinates of the forming area of ​​the irregular sheet metal part, the numerical defect features and the image defect features are weighted and fused to obtain the second forming quality feature.

7. The method for correcting the progressive multi-point forming process of irregularly shaped sheet metal parts according to claim 6, characterized in that, The process of obtaining multi-dimensional matching results based on the suspected defect areas of the irregularly shaped sheet metal parts, and determining the suspected associated influence areas, includes: Based on the suspected defect areas of the irregular sheet metal parts, determine multi-dimensional matching rules; The multi-dimensional matching rules are matched with the forming areas of the irregular sheet metal parts respectively to obtain the multi-dimensional matching results of the suspected defect areas; The irregular sheet metal forming area where the multi-dimensional matching result exceeds the set matching degree threshold is identified as a suspicious correlation influence area.

8. The method for correcting the progressive multi-point forming process of irregularly shaped sheet metal parts according to claim 1, characterized in that, The step of merging the suspected defect areas and the suspected associated influence areas of the irregularly shaped sheet metal part into a target defect area set includes: Obtain the spatial coordinates, defect type, and defect severity information of the suspected defect area, and establish a suspected defect area information set; Obtain the spatial coordinates of the suspected associated influence area and the matching dimension information with the suspected defect area, and establish a set of information on the suspected associated influence area; The set of information on suspected defective regions and the set of information on suspected associated regions are deduplicated and integrated to form a set of target defective regions.

9. The method for correcting the progressive multi-point forming process of irregularly shaped sheet metal parts according to claim 1, characterized in that, The step of outputting a forming process correction scheme based on the target defect region set and the multi-dimensional matching results includes: The attribute information of the forming area of ​​each irregular sheet metal part in the target defect area set is analyzed; wherein, the attribute information includes spatial coordinates, defect type and defect severity information; Based on the spatial coordinates, defect type, and defect severity information, combined with the multi-dimensional matching results, the cause of defects corresponding to each target defect region in the target defect region set is located. Differentiated process correction strategies are generated for different causes of the aforementioned defects; The process correction strategy is associated with the corresponding target defect region to generate a forming process correction scheme.

10. A progressive multi-point forming process correction system for irregularly shaped sheet metal parts, characterized in that, include: The acquisition unit is used to acquire the forming parameters of the irregular sheet metal part; wherein, the forming parameters include basic data of the blank, process parameter data and equipment status data; The combination unit is used to determine multiple low-risk forming combinations based on the billet basic data, the process parameter data, and the equipment status data; A defect unit is used to identify suspected defect areas of irregularly shaped sheet metal parts based on a variety of the aforementioned low-risk forming combinations. The association unit is used to obtain multi-dimensional matching results based on the suspected defect area of ​​the irregular sheet metal part, and determine the suspected associated influence area; wherein, the suspected associated influence area refers to other forming areas that are affected by the defect in the suspected defect area; the multi-dimensional matching result refers to the result obtained by matching the parameter similarity between the local area of ​​the irregular sheet metal part and the suspected defect area from three dimensions: basic data of the blank, process parameter data, and equipment status data; The target unit is used to merge the suspected defect area and the suspected associated influence area of ​​the irregular sheet metal part into a target defect area set; The forming unit is used to output a forming process correction scheme based on the target defect area set and the multi-dimensional matching result; wherein the forming process correction scheme is a forming scheme that requires secondary adjustment of forming parameters during the low-risk forming combination forming process of irregular sheet metal parts.

Images