Body-in-white door opening cmm quality data visualization traceability method, system and device

Abstract

This invention discloses a method, system, and device for visualizing and tracing CMM quality data of body-in-white door openings. The method includes: acquiring CMM inspection data from multiple inspection points in the body-in-white door opening area; performing heterogeneous tolerance normalization processing on the measured deviation values ​​of each inspection point according to the tolerance type; generating a spatial pattern signature based on the normalized deviation values ​​to characterize the overall deviation pattern of the body-in-white door openings; calculating the quality severity index of the corresponding body based on the normalized deviation values ​​of each inspection point; calculating the anomaly clustering degree based on the spatial pattern signature and the quality severity index; and mapping the data within the abnormal time period to a digital twin model for visual playback based on the anomaly clustering degree. The method, system, and device for visualizing and tracing CMM quality data of body-in-white door openings of this invention can uniformly evaluate heterogeneous tolerance measurement points, automatically identify abnormal spatial patterns, focus on high-risk production time periods, and achieve visual traceability by combining with a digital twin model.

Description

Technical Field

[0001] This invention relates to the field of body-in-white inspection technology, and more specifically, to a method, system, and device for visualizing and tracing CMM quality data of body-in-white door openings, particularly to a method, system, electronic device, and computer-readable storage medium for visualizing and tracing CMM quality data of body-in-white door openings. Background Technology

[0002] The welding quality of the automotive body-in-white directly affects the vehicle's sealing performance, assembly accuracy, and NVH (noise, vibration, and harshness) performance. The sealing and fitting areas of door openings, with their complex spatial geometry and dense assembly constraints, are prone to accumulating localized deviations during welding, leading to various quality issues such as abnormal door closing force, increased wind noise, and water leakage. Currently, production lines typically use coordinate measuring machines (CMMs) for offline inspection of key measuring points in the door opening areas of the body-in-white to obtain dimensional deviation data for these points.

[0003] In existing technologies, the output of CMM inspection results is a static inspection report presented in tabular form. Although it can reflect the dimensional status of a single vehicle body under a certain inspection, the characteristics of this type of data are discrete and two-dimensional, making it difficult to continuously correlate the inspection results with production cycle, tooling status, and historical batches. Consequently, it is difficult to quickly trace products with abnormal quality events on the manufacturing site.

[0004] Different inspection points in the body-in-white doorway area often have different assembly functions and tolerance constraints in their design. Some inspection points use symmetrical tolerance evaluation methods, while others use unilateral tolerance evaluation methods. Existing quality analysis methods usually directly compare the original deviation values ​​or use a uniform threshold strategy for early warning, which makes it difficult to eliminate the differences in evaluation scales caused by heterogeneous tolerances, making it difficult to conduct cross-sectional unified analysis between different inspection points.

[0005] In complex, multi-station welding scenarios, single-point deviations appear sporadic, but combinations of spatial deviations from multiple points are more likely to reflect specific process issues, such as fixture wear, locating pin drift, abnormal threshold support, or welding clamping force imbalance. Existing methods typically lack structured characterization techniques for the combined state of multiple measurement points, and also lack identification mechanisms for high-incidence periods of anomalies in continuous production batches. Under existing technologies, troubleshooting on the production floor still heavily relies on manual experience.

[0006] Therefore, there is an urgent need for a method, system, and equipment for visualizing and tracing CMM quality data of body-in-white door openings. Summary of the Invention

[0007] The purpose of this invention is to provide a method, system, and device for visualizing and tracing CMM quality data of body-in-white door openings, in order to solve the problems in the prior art. It can uniformly evaluate heterogeneous tolerance measurement points, automatically identify abnormal spatial patterns, focus on high-risk production time periods, and achieve visual traceability by combining digital twin models.

[0008] This invention provides a method for visualizing and tracing CMM quality data of body-in-white door openings, comprising:

[0009] Acquire CMM inspection data from multiple inspection points in the doorway area of ​​the white body, wherein the CMM inspection data includes at least the body markings, measurement time, and the measured deviation value corresponding to each inspection point;

[0010] Heterogeneous tolerance normalization is performed on the measured deviation values ​​of the corresponding test points according to the tolerance type of each test point. The tolerance types of the test points include: symmetrical tolerance test points and unilateral tolerance test points. The symmetrical tolerance test points are normalized according to the corresponding tolerance limit, and the unilateral tolerance test points are normalized according to the corresponding unilateral tolerance upper limit. When the measured deviation value is in the prohibited deviation direction, a penalty mapping process is performed to obtain the normalized deviation value of each test point.

[0011] The corresponding discrete state symbol is determined based on the normalized deviation value of each detection point, and the discrete state symbol is combined according to the preset detection point order to generate a spatial pattern signature for characterizing the overall deviation shape of the white body door opening.

[0012] The quality severity index of the corresponding vehicle body is calculated based on the normalized deviation value of each detection point, and the quality event objects, which include at least vehicle identification, measurement time, normalized deviation value, spatial pattern signature and quality severity index, are arranged in the order of measurement time to form a quality digital thread;

[0013] In the quality digital thread, the frequency of occurrence of target signatures is counted according to the sliding time window, and the abnormal clustering degree is calculated in combination with the quality severity index within the sliding time window;

[0014] When the abnormal clustering degree exceeds a preset threshold, the corresponding abnormal time period is determined, and the spatial pattern signature and normalized deviation value within the abnormal time period are mapped to virtual semantic anchor points in the digital twin model of the white body door opening that correspond one-to-one with each detection point for visualization playback.

[0015] The CMM quality data visualization and traceability method for the white body door opening described above preferably includes at least the following detection points: hinge position measurement point on the upper part of the door opening, door lock hook position measurement point, hinge position measurement point on the lower part of the door opening, and sill position measurement point.

[0016] The acquisition of CMM inspection data for multiple inspection points in the white body door opening area includes:

[0017] The measured deviation value is obtained from the coordinate measuring machine, and the vehicle body markings and production-related information are obtained from the production execution system.

[0018] The above-described method for visualizing and tracing CMM quality data of body-in-white door openings, wherein, preferably, the...

[0019] The symmetrical tolerance measurement points are normalized according to the corresponding tolerance limits, specifically by calculating the normalized deviation value using the following formula:

[0020] ,

[0021] in, This represents the normalized deviation value. Indicates the first The car body is in the first The measured deviation value of each measuring point Indicates the first Symmetrical tolerance limits for each measuring point;

[0022] The process of normalizing the unilateral tolerance measurement points according to the corresponding unilateral tolerance upper limit, and performing penalty mapping processing when the measured deviation value is in the prohibited deviation direction, to obtain the normalized deviation value of each detection point, specifically includes:

[0023] If the measured deviation value is in the permissible direction, the normalized deviation value is calculated using the following formula:

[0024] ,

[0025] in, Indicates the first The upper limit of the single-sided tolerance for each measuring point, with an allowable deviation range of [0, ...]. ], ≥0 indicates an allowed direction. <0 indicates a prohibited direction;

[0026] If the measured deviation value is in the prohibited direction, the normalized deviation value is calculated using the penalty mapping function, which is expressed as:

[0027] ,

[0028] in, It is represented as the negative deviation extreme value penalty factor, λ>0.

[0029] The above-described method for visualizing and tracing CMM quality data of body-in-white door openings, preferably, involves determining the corresponding discrete state symbols based on the normalized deviation values ​​of each detection point, and combining the discrete state symbols according to a preset detection point order to generate a spatial pattern signature characterizing the overall deviation morphology of the body-in-white door openings, including:

[0030] The corresponding discrete state symbol is determined based on the normalized deviation value of each detection point, and the spatial pattern signature is formed by combining the symbols according to the preset detection point order. ;

[0031] The status level of each detection point is determined based on the absolute value of the normalized deviation value of each detection point, wherein the status level includes at least normal status, warning status and out-of-tolerance status.

[0032] Based on the comparison between the normalized deviation value of each detection point and the preset discrete threshold, the discrete state symbol corresponding to each detection point is determined. Specifically, when the absolute value of the normalized deviation value is not greater than the preset discrete threshold, the discrete state symbol is a normal symbol; when the normalized deviation value is greater than the preset discrete threshold, the discrete state symbol is a positive deviation symbol; and when the normalized deviation value is less than the negative of the preset discrete threshold, the discrete state symbol is a negative deviation symbol.

[0033] The spatial pattern signature is formed by combining the discrete state symbols corresponding to each detection point according to the preset detection point sequence.

[0034] In the above-described method for visualizing and tracing CMM quality data of body-in-white door openings, preferably, the quality severity index is used to characterize the severity of the overall quality of the vehicle body.

[0035] The process involves calculating the corresponding vehicle body quality severity index based on the normalized deviation value of each detection point, and arranging quality event objects, including at least vehicle identification, measurement time, normalized deviation value, spatial pattern signature, and quality severity index, in chronological order to form a quality digital thread, including:

[0036] The quality severity index of the corresponding vehicle body is calculated based on the normalized deviation value of each detection point. The quality severity index is the maximum value among the absolute values ​​of the normalized deviation values ​​corresponding to multiple detection points, and is expressed by the following formula:

[0037] ;

[0038] After obtaining the normalized deviation value, spatial pattern signature, and quality severity index, the detection results of each vehicle body are encapsulated into a quality event object. The quality event object includes: vehicle body identification, measurement time, normalized deviation vector, quality severity index, and spatial pattern signature.

[0039] Multiple quality event objects are arranged in chronological order to form a quality digital thread corresponding to the doorway area of ​​the white body.

[0040] The above-described method for visualizing and tracing CMM quality data of body-in-white door openings, preferably, involves, within the quality data thread, statistically analyzing the frequency of target signature occurrences according to a sliding time window, and calculating the anomaly clustering degree in conjunction with the quality severity index within the sliding time window, including:

[0041] In the aforementioned quality digital thread, time windows are divided according to a preset time span or a preset number of vehicles. Spatial pattern signatures of continuously produced vehicles are statistically analyzed using a sliding time window. Anomaly clustering is calculated by combining the frequency of target signature occurrences and the quality severity index.

[0042] The degree of anomaly clustering is determined by combining the frequency percentage of the target signature within the sliding time window and the statistical results of the severity indices of each vehicle quality within the sliding time window. Specifically, this includes:

[0043] Let the current sliding time window be W(t,Δt), which contains N vehicle bodies; the number of times a certain target signature appears within this time window is... The abnormal clustering degree is calculated using the following formula. :

[0044]

[0045] in, Indicates the number of vehicles. This indicates the frequency of the label name within the current time window. This indicates the severity index of the quality. This indicates the overall severity level of the current time window.

[0046] The above-described method for visualizing and tracing CMM quality data of body-in-white door openings, preferably, involves mapping data within the abnormal time period to a digital twin model for visualization playback based on the anomaly clustering degree, including:

[0047] If the abnormal clustering degree exceeds a preset threshold, the corresponding abnormal time period is determined, and the spatial pattern signature and normalized deviation value within the abnormal time period are mapped to the digital twin model of the body-in-white doorway. The virtual semantic anchor points are then visualized and replayed. The virtual semantic anchor points are preset in the digital twin model of the body-in-white doorway and correspond one-to-one with each of the detection points. The visualization and replay includes color mapping display of each virtual semantic anchor point based on the normalized deviation value.

[0048] Establish a pre-defined association table to characterize the relationship between spatial pattern signatures and process objects;

[0049] Based on the preset association table, investigation clues are generated.

[0050] The present invention also provides a visual traceability system for CMM quality data of body-in-white door openings using the above method, comprising:

[0051] The data acquisition module is used to acquire CMM inspection data from multiple inspection points in the doorway area of ​​the white body. The CMM inspection data includes at least the body markings, measurement time, and the measured deviation value corresponding to each inspection point.

[0052] The heterogeneous tolerance normalization module is used to perform heterogeneous tolerance normalization processing on the measured deviation values ​​of the corresponding detection points according to the tolerance type of each detection point, so as to obtain the normalized deviation value of each detection point. The tolerance types of the detection points include: symmetrical tolerance measurement points and unilateral tolerance measurement points. The symmetrical tolerance measurement points are normalized according to the corresponding tolerance limit, and the unilateral tolerance measurement points are normalized according to the corresponding unilateral tolerance upper limit. When the measured deviation value is in the prohibited deviation direction, a penalty mapping process is performed to obtain the normalized deviation value of each detection point.

[0053] The pattern signature generation module is used to determine the corresponding discrete state symbols based on the normalized deviation values ​​of each detection point, and to combine the discrete state symbols according to the preset detection point order to generate a spatial pattern signature that characterizes the overall deviation shape of the white body door opening.

[0054] The Quality Digital Thread Construction Module is used to calculate the corresponding vehicle body quality severity index based on the normalized deviation value of each detection point, and to arrange quality event objects, including at least vehicle body markings, measurement time, normalized deviation value, spatial pattern signature and quality severity index, in the order of measurement time to form a quality digital thread;

[0055] The clustering analysis module is used to count the frequency of occurrence of target signatures according to a sliding time window in the quality digital thread, and to calculate the abnormal clustering degree in combination with the quality severity index within the sliding time window.

[0056] The visualization traceability module is used to determine the corresponding abnormal time period when the abnormal clustering degree exceeds a preset threshold, and to map the spatial pattern signature and normalized deviation value within the abnormal time period to virtual semantic anchor points in the digital twin model of the white body door opening that correspond one-to-one with each detection point, so as to perform visualization playback and output the investigation clues of the corresponding tooling or fixture.

[0057] The present invention also provides an electronic device, including a processor and a memory, wherein the memory and the processor are connected; the memory is used to store a computer program; and the processor calls the computer program stored in the memory to execute the above-described method.

[0058] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a computer, performs the above-described method.

[0059] This invention provides a method, system, and device for visualizing and tracing CMM quality data of body-in-white door openings. Addressing the pain points of CMM inspection data for body-in-white door openings, it unifies the deviation values ​​of different types of measuring points into dimensionless indices through heterogeneous tolerance normalization processing. Then, it uses spatial pattern signatures to represent the overall deviation pattern and calculates the anomaly clustering degree in conjunction with a quality severity index to automatically identify abnormal production periods. Finally, it visualizes and replays the quality status through a digital twin model, outputting clues for tooling, fixtures, and other troubleshooting methods. This achieves unified evaluation of heterogeneous tolerance measuring points, automatic identification of abnormal spatial patterns, focusing on high-risk production periods, and visualized traceability through a digital twin model. It breaks down the evaluation barriers of heterogeneous data, reduces reliance on human experience, and significantly improves the efficiency and accuracy of quality traceability, providing data support for quality control in the production process. It solves the technical problems of existing technologies, such as discrete storage of CMM inspection data for body-in-white door openings, difficulty in unified evaluation of heterogeneous tolerances, difficulty in automatic identification of abnormal patterns, reliance on human experience in the traceability process, and excessively large investigation scope. Attached Figure Description

[0060] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described below with reference to the accompanying drawings, wherein:

[0061] Figure 1 A flowchart illustrating an embodiment of the CMM quality data visualization and traceability method for body-in-white door openings provided by the present invention;

[0062] Figure 2 A logic diagram of an embodiment of the CMM quality data visualization and traceability method for body-in-white door openings provided by the present invention;

[0063] Figure 3 A flowchart illustrating the process of heterogeneous tolerance normalization.

[0064] Figure 4 A schematic diagram of the architecture of a digital twin model;

[0065] Figure 5 This is a schematic diagram showing the distribution of semantic measurement points and the binding of semantic anchor points for the door openings of a white car body.

[0066] Figure 6 This is a structural block diagram of an embodiment of the CMM quality data visualization and traceability system for body-in-white door openings provided by the present invention. Detailed Implementation

[0067] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. The descriptions of the exemplary embodiments are merely illustrative and are in no way intended to limit the present disclosure or its application or use. The present disclosure may be implemented in many different forms and is not limited to the embodiments described herein. These embodiments are provided so that the present disclosure will be thorough and complete, and will fully express the scope of the disclosure to those skilled in the art. It should be noted that, unless specifically stated otherwise, the relative arrangement of components and steps, the composition of materials, numerical expressions, and values ​​set forth in these embodiments should be interpreted as exemplary only and not as limiting.

[0068] The terms “first,” “second,” and similar terms used in this disclosure do not indicate any order, quantity, or importance, but are merely used to distinguish different parts. Terms such as “including” or “contains” mean that the element preceding the term encompasses the element listed after it, and do not exclude the possibility of encompassing other elements as well. Terms such as “above” and “below” are used only to indicate relative positional relationships; when the absolute position of the described object changes, this relative positional relationship may also change accordingly.

[0069] In this disclosure, when a specific component is described as being located between a first component and a second component, an intermediary component may or may not be present between the specific component and the first or second component. When a specific component is described as connecting to other components, the specific component may be directly connected to the other components without having an intermediary component, or it may not be directly connected to the other components but may have an intermediary component.

[0070] All terms used in this disclosure (including technical or scientific terms) have the same meaning as understood by one of ordinary skill in the art to which this disclosure pertains, unless otherwise specifically defined. It should also be understood that terms defined in a general dictionary, such as a dictionary, should be interpreted as having a meaning consistent with their meaning in the context of the relevant art, and not as having an idealized or highly formalized meaning, unless expressly defined herein.

[0071] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, they should be considered part of the specification.

[0072] like Figure 1 and Figure 2 As shown, the visual traceability method for CMM quality data of body-in-white door openings provided in this embodiment includes the following steps in actual execution: Step S101: Obtain CMM detection data of multiple detection points in the body-in-white door opening area. The CMM detection data includes at least: Body ID, measurement time, and the measured deviation value corresponding to each detection point.

[0073] Specifically, the measured deviation value is obtained from the coordinate measuring machine, and the vehicle body markings and production-related information are obtained from the production execution system.

[0074] The vehicle identification mark is used to uniquely identify the corresponding physical vehicle body, and the measurement time is used to characterize the moment of detection and construct a continuous time axis, with the measured deviation value corresponding to each detection point. Further, in one embodiment of the invention, the CMM detection data also includes metadata such as part number, measurement direction, theoretical value, and tolerance limit.

[0075] Furthermore, in one embodiment of the present invention, the detection points include at least: a hinge position measuring point on the door opening, a door lock hook position measuring point, a hinge position measuring point below the door opening, and a sill position measuring point. It should be noted that the present invention does not specifically limit the location of the detection points. The body-in-white door opening CMM quality data visualization and traceability method provided by the present invention is applied to the dimensional quality traceability scenario of the body-in-white door opening area in an automotive welding workshop. The body-in-white door opening area can be a rear door opening area, correspondingly setting multiple measuring points with assembly semantics. For example, according to actual application requirements, such as... Figure 5 As shown, the detection points include: the upper hinge position measuring point P1 of the rear door opening, the rear door lock hook position measuring point P2, the lower hinge position measuring point P3 of the rear door opening, and the sill position measuring point P4 of the rear door opening.

[0076] Through the data acquisition method in step S101, the original discrete CMM data can be organized into basic data objects for subsequent traceability calculations.

[0077] Step S102: Perform heterogeneous tolerance normalization processing on the measured deviation values ​​of the corresponding detection points according to the tolerance type of each detection point. The tolerance types of the detection points include: symmetrical tolerance measurement points and unilateral tolerance measurement points. The symmetrical tolerance measurement points are normalized according to the corresponding tolerance limit, and the unilateral tolerance measurement points are normalized according to the corresponding unilateral tolerance upper limit. When the measured deviation value is in the prohibited deviation direction, a penalty mapping process is performed to obtain the normalized deviation value of each detection point.

[0078] Since different measurement points may have different tolerance constraint types in engineering, this invention employs different normalization methods for different measurement points. Specifically, for example... Figure 3 As shown, for symmetrical tolerance measurement points (e.g., P1, P2, and P3), normalization according to the corresponding tolerance limits is performed, specifically including calculating the normalized deviation value using the following formula:

[0079] ,

[0080] in, This represents the normalized deviation value. Indicates the first The car body is in the first The measured deviation value of each measuring point Indicates the first Symmetrical tolerance limits for each measuring point.

[0081] For single-sided tolerance measurement points (e.g., P4), normalization is performed according to the corresponding single-sided tolerance upper limit, and when the measured deviation value is in the prohibited deviation direction, a penalty mapping process is applied, specifically including:

[0082] If the measured deviation value is in the permissible direction, the normalized deviation value is calculated using the following formula:

[0083] ,

[0084] in, Indicates the first The upper limit of the single-sided tolerance for each measuring point, and the allowable deviation range are... , ≥0 indicates an allowed direction. <0 indicates the prohibited direction, meaning the normalization is performed according to the ratio of the measured deviation value to the upper limit of the unilateral tolerance.

[0085] If the measured deviation value is in the prohibited direction, the normalized deviation value is calculated using the penalty mapping function, which is expressed as:

[0086] ,

[0087] in, It is represented as the negative deviation extreme value penalty factor, λ>0.

[0088] By using the penalty mapping function, the normalized result corresponding to the prohibited direction deviation can be numerically inferior to the allowable direction deviation result, highlighting its engineering deterioration attribute. Through this setting, the prohibited direction deviation can be directly distinguished from the normal deviation state after normalization and given higher weight in subsequent pattern recognition.

[0089] In one embodiment of the present invention, such as Figure 3 As shown, input the CMM measured deviation data (vehicle number, measurement time, measurement point number (P1-P4) and the corresponding measured deviation value). Then, identify the tolerance type of the measuring point and determine whether it is a symmetrical tolerance measuring point; if it is a symmetrical tolerance measuring point, normalize it according to the symmetrical tolerance and output the normalized deviation value. If the measurement point has an asymmetric tolerance, determine whether the measured deviation is in an allowable direction. If it is in an allowable direction, normalize it according to the upper limit of the single-sided tolerance and output the normalized deviation value. If it is in a prohibited direction, perform negative deviation penalty mapping and output the normalized deviation value. The output normalized deviation value is used for state classification, pattern signature extraction, and anomaly clustering analysis.

[0090] Step S102 allows the measured deviation values ​​under different tolerance types to be uniformly mapped to dimensionless normalized deviation values.

[0091] Step S103: Determine the corresponding discrete state symbol based on the normalized deviation value of each detection point, and combine the discrete state symbols according to the preset detection point order to generate a spatial pattern signature for characterizing the overall deviation shape of the white body door opening.

[0092] The spatial pattern signature is used to characterize the overall deviation shape of the door opening of the white body.

[0093] In one embodiment of the CMM quality data visualization and traceability method for body-in-white door openings of the present invention, step S103 may specifically include:

[0094] Step S1031: Determine the status level of the corresponding detection point based on the absolute value of the normalized deviation value of each detection point.

[0095] The status levels include at least normal (OK), warning (WARN), and out-of-tolerance (NG). In this invention, the status level of the detection point is used to determine whether the status of each measurement point is normal, warning, or out of tolerance, and then different colors are displayed in the subsequent visualization playback step according to different status levels.

[0096] Step S1032: Determine the discrete state symbol corresponding to each detection point based on the comparison result between the normalized deviation value of each detection point and the preset discrete threshold.

[0097] Wherein, when the absolute value of the normalized deviation is not greater than the preset discrete threshold, the discrete state sign is a normal sign; when the normalized deviation is greater than the preset discrete threshold, the discrete state sign is a positive deviation sign; when the normalized deviation is less than the negative of the preset discrete threshold, the discrete state sign is a negative deviation sign. The correspondence between the discrete state sign and the normalized deviation value is expressed by the following formula:

[0098]

[0099] in, Representing discrete state symbols, Represents the discrete threshold of the state. This represents the normalized deviation value. In one embodiment of the invention, the state discrete threshold... A value of 0.8 is used so that once a measuring point enters a warning or out-of-tolerance state, its deviation direction will be reflected in the spatial pattern signature. It should be noted that this invention uses a state discrete threshold. The value of is not specifically limited.

[0100] Step S1033: Combine the discrete state symbols corresponding to each detection point according to the preset detection point sequence to form the spatial pattern signature.

[0101] This invention employs a two-layer encoding. The first layer encodes the state level of the detection point, which is used for color-coded display in subsequent visualization playback steps. The second layer encodes the discrete state symbols of the detection points, used to characterize the deviation direction of each detection point. In one embodiment of this invention, the state levels and spatial pattern signatures of the four measuring points P1-P4 are shown in Table 1. As shown in Table 1, although both positive warnings and positive deviations are displayed as "+" in the spatial pattern signature, their degree of abnormality can be determined by combining the state level.

[0102] Table 1. Status level and spatial mode signature of each measuring point

[0103]

[0104] In this invention, spatial pattern signatures are used to characterize the combination of spatial deviation directions of multiple detection points, while the severity of each detection point is recorded separately by its state level and normalized deviation value. Normal, positive deviation, and negative deviation symbols can be represented as "0", "+", and "-", respectively. For example, when the discrete state symbols of the four detection points are "0, 0, 0, +", the corresponding spatial pattern signature is "000+", indicating that the anomaly is mainly concentrated in the region corresponding to the fourth detection point. The degree of anomaly at the fourth detection point can be further determined by combining its state level and normalized deviation value.

[0105] Step S104: Calculate the corresponding vehicle body quality severity index based on the normalized deviation value of each detection point, and arrange the quality event objects, including at least vehicle body markings, measurement time, normalized deviation value, spatial pattern signature, and quality severity index, in the order of measurement time to form a quality digital thread.

[0106] The quality severity index is used to characterize the severity of the overall vehicle body quality, facilitating the quantification of the overall quality deviation level of a single vehicle body. In one embodiment of the body-in-white doorway CMM quality data visualization and traceability method of the present invention, step S104 may specifically include:

[0107] Step S1041: Calculate the corresponding vehicle body quality severity index based on the normalized deviation value of each detection point. The quality severity index is the maximum value among the absolute values ​​of the normalized deviation values ​​corresponding to multiple detection points, expressed by the following formula:

[0108] .

[0109] That is, the maximum value among the normalized absolute values ​​of multiple measurement points on the current vehicle body is taken as the quantitative indicator of the severity of the overall quality of the vehicle body.

[0110] Step S1042: After obtaining the normalized deviation value, spatial pattern signature, and quality severity index, the detection results of each vehicle body are encapsulated into a quality event object. The quality event object includes: vehicle body identification, measurement time, normalized deviation vector, quality severity index, and spatial pattern signature.

[0111] The quality event objects include: vehicle identification, measurement time, normalized deviation vector, quality severity index, and spatial pattern signature.

[0112] Step S1043: After arranging multiple quality event objects in chronological order, a quality digital thread corresponding to the white body doorway area is formed.

[0113] The use of quality digital threads facilitates subsequent continuous traceability and time window analysis.

[0114] Step S105: In the quality digital thread, the frequency of occurrence of the target signature is counted according to the sliding time window, and the abnormal clustering degree is calculated in combination with the quality severity index within the sliding time window.

[0115] In the quality digital thread, time windows are divided according to a preset time span or a preset number of vehicles. The spatial pattern signatures of continuously produced vehicles are statistically analyzed by sliding time windows. The abnormal clustering degree is calculated by combining the frequency of occurrence of the target signature and the quality severity index.

[0116] By calculating the anomaly clustering degree, persistent or intermittent anomalies can be identified. Specifically, the anomaly clustering degree is determined by combining the frequency percentage of the target signature within the sliding time window and the statistical results of each vehicle quality severity index within the sliding time window.

[0117] Specifically, let the current sliding time window be W(t,Δt), which contains N vehicle bodies; the number of times a certain target signature appears within this time window is... The abnormal clustering degree is calculated using the following formula. :

[0118]

[0119] in, Indicates the number of vehicles. This indicates the frequency of the label name within the current time window. This indicates the severity index of the quality. This indicates the overall severity level of the current time window. Through the above coupling method, both the recurrence characteristics of a certain type of anomaly pattern and the intensity of quality deterioration can be reflected simultaneously.

[0120] Step S106: When the abnormal clustering degree exceeds the preset threshold, determine the corresponding abnormal time period, and map the spatial pattern signature and normalized deviation value within the abnormal time period to the virtual semantic anchor point in the digital twin model of the white body door opening that corresponds one-to-one with each detection point for visualization playback.

[0121] In one embodiment of the CMM quality data visualization and traceability method for body-in-white door openings of the present invention, step S106 may specifically include:

[0122] Step S1061: If the abnormal clustering degree exceeds the preset threshold, the corresponding abnormal time period is determined, and the spatial pattern signature and normalized deviation value within the abnormal time period are mapped to the digital twin model of the white body doorway, and the virtual semantic anchor point is visualized and replayed.

[0123] When the abnormal clustering degree corresponding to a certain target signature is detected to exceed a preset threshold, the production stage corresponding to the current sliding time window is determined to be a period of high incidence of abnormalities. In specific implementation, when the abnormal clustering degree exceeds the threshold, the system extracts the quality event objects within the corresponding abnormal time period and maps the normalized deviation value and spatial pattern signature within that time period to the digital twin model of the body-in-white door opening. Conversely, if the abnormal clustering degree does not exceed the preset threshold, the system returns to the step of collecting CMM detection data, i.e., returns to step S101.

[0124] In this example, the digital twin model is a lightweight 3D mesh model of the doorway area. The virtual semantic anchor points are pre-defined in the digital twin model of the doorway of the vehicle body and correspond one-to-one with each of the detection points, such as... Figure 5 As shown, the virtual semantic anchors are A1-A4, corresponding to P1-P4. (As...) Figure 4 As shown, the architecture of the digital twin model includes: a data governance layer, a twin model layer, a traceability algorithm layer, and an application service layer. In its implementation, the underlying quality data is constructed using the vehicle identification number (VIN) and measurement time as the primary key, and then input into the digital twin model. The data governance layer is used for collecting CMM measurement data and MES production data, extracting VINs, data cleaning, and structured reorganization. The twin model layer is used for constructing a lightweight 3D model of the body-in-white's door openings, semantic anchor mapping, binding CMM physical and virtual measurement points, and determining key measurement points for door openings. The traceability algorithm layer is the core algorithm layer, used for heterogeneous tolerance normalization, quality event classification, spatial pattern signature extraction, time window clustering analysis, and identification of high-incidence anomaly segments. The application service layer is used for 3D quality status visualization, historical defect event playback, root cause clue focusing, and tooling / fixture inspection guidance.

[0125] In one embodiment of the present invention, the visualization playback includes performing color mapping display on each of the virtual semantic anchors according to the normalized deviation value. As mentioned above, green, yellow, and red are used to represent normal, warning, and out-of-tolerance states, respectively. In one embodiment of the present invention, the mapping relationship between state level and color is shown in Table 2. As shown in Table 2, when the state level is normal (OK), the display color is green; when the warning state is WARN, the display color is yellow; and when the state level is out-of-tolerance (NG), the display color is red. The relationship between the state level and the absolute value of the normalized deviation value can be expressed by the following formula:

[0126]

[0127] in, Indicates the status level of the detection point. This indicates that the status level is normal. This indicates that the status level is a warning state. This indicates that the state level is out of tolerance. This represents the normalized deviation value.

[0128] Table 2. Mapping Relationship between Status Level and Color

[0129]

[0130] In another embodiment of the present invention, for target measurement points with persistent anomalies, the display effect can be enhanced by methods such as flashing, magnification, or highlighting the border. Through these methods, field engineers can dynamically replay the quality status during the abnormal time period along the timeline, intuitively observe whether the anomaly is concentrated in the threshold area, hinge area, or hook area, and combine this with the corresponding pattern signature to output the investigation targets such as the threshold support block, positioning fixture, welding robot trajectory, or clamping force parameters, thereby improving the efficiency of root cause localization.

[0131] Step S1062: Establish a preset association table to characterize the relationship between spatial pattern signatures and process objects.

[0132] Step S1063: Generate investigation clues based on the preset association table.

[0133] Specifically, the abnormal pattern signature is matched with at least one of the candidate tooling, fixture, welding robot trajectory, and process parameters to generate investigation clues.

[0134] Furthermore, such as Figure 6 As shown, the present invention also provides a visual traceability system for CMM quality data of body-in-white door openings using the above method, comprising:

[0135] Data acquisition module 1 is used to acquire CMM inspection data of multiple inspection points in the door opening area of ​​the white body. The CMM inspection data includes at least the body markings, measurement time, and the measured deviation value corresponding to each inspection point.

[0136] Heterogeneous tolerance normalization module 2 is used to perform heterogeneous tolerance normalization processing on the measured deviation value of the corresponding detection point according to the tolerance type of each detection point, so as to obtain the normalized deviation value of each detection point. The tolerance type of the detection point includes: symmetrical tolerance measurement point and unilateral tolerance measurement point. The symmetrical tolerance measurement point is normalized according to the corresponding tolerance limit, and the unilateral tolerance measurement point is normalized according to the corresponding unilateral tolerance upper limit. When the measured deviation value is in the prohibited deviation direction, penalty mapping processing is performed to obtain the normalized deviation value of each detection point.

[0137] The pattern signature generation module 3 is used to determine the corresponding discrete state symbol based on the normalized deviation value of each detection point, and combine the discrete state symbols according to the preset detection point order to generate a spatial pattern signature that characterizes the overall deviation shape of the white body door opening.

[0138] The Quality Digital Thread Construction Module 4 is used to calculate the corresponding vehicle body quality severity index based on the normalized deviation value of each detection point, and to arrange quality event objects, including at least vehicle body markings, measurement time, normalized deviation value, spatial pattern signature and quality severity index, in the order of measurement time to form a quality digital thread;

[0139] Clustering analysis module 5 is used to count the frequency of occurrence of target signatures according to a sliding time window in the quality digital thread, and to calculate the abnormal clustering degree in combination with the quality severity index within the sliding time window;

[0140] The visualization traceability module 6 is used to determine the corresponding abnormal time period when the abnormal clustering degree exceeds the preset threshold, and to map the spatial pattern signature and normalized deviation value within the abnormal time period to the virtual semantic anchor point in the digital twin model of the white body door opening that corresponds one-to-one with each detection point, so as to perform visualization playback and output the investigation clues of the corresponding tooling or fixture.

[0141] Furthermore, the present invention also provides an electronic device, including a processor and a memory, wherein the memory and the processor are connected; the memory is used to store a computer program; the processor invokes the computer program stored in the memory to execute the above-described method.

[0142] Furthermore, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a computer, performs the above-described method.

[0143] The present invention provides a visual traceability method, system, and device for CMM quality data visualization of body-in-white door openings. Addressing the pain points of CMM inspection data for body-in-white door openings, it unifies the deviation values ​​of different types of measuring points into dimensionless indices through heterogeneous tolerance normalization processing. Then, it uses spatial pattern signatures to represent the overall deviation pattern, calculates the anomaly clustering degree in conjunction with the quality severity index, and automatically identifies abnormal production periods. Finally, it visualizes and replays the quality status through a digital twin model, outputting clues for tooling, fixtures, and other troubleshooting methods. This achieves unified evaluation of heterogeneous tolerance measuring points, automatic identification of abnormal spatial patterns, focusing on high-risk production periods, and visual traceability through a digital twin model. It breaks down the evaluation barriers of heterogeneous data, reduces reliance on human experience, and significantly improves the efficiency and accuracy of quality traceability, providing data support for quality control in the production process. It solves the technical problems of discrete storage of CMM inspection data for body-in-white door openings, difficulty in unified evaluation of heterogeneous tolerances, difficulty in automatic identification of abnormal patterns, reliance on human experience in the traceability process, and excessively large investigation scope in existing technologies.

[0144] The embodiments of this disclosure have now been described in detail. To avoid obscuring the concept of this disclosure, some details known in the art have not been described. Those skilled in the art can fully understand how to implement the technical solutions disclosed herein based on the above description.

[0145] While specific embodiments of this disclosure have been described in detail by way of examples, those skilled in the art should understand that the examples are for illustrative purposes only and not intended to limit the scope of this disclosure. Those skilled in the art should understand that modifications can be made to the above embodiments or equivalent substitutions can be made to some technical features without departing from the scope and spirit of this disclosure. The scope of this disclosure is defined by the appended claims.

Claims

1. A method of white body door opening CMM quality data visualization traceability, characterized in that, include: Acquire CMM inspection data from multiple inspection points in the doorway area of ​​the white body, wherein the CMM inspection data includes at least the body markings, measurement time, and the measured deviation value corresponding to each inspection point; Heterogeneous tolerance normalization is performed on the measured deviation values ​​of the corresponding test points according to the tolerance type of each test point. The tolerance types of the test points include: symmetrical tolerance test points and unilateral tolerance test points. The symmetrical tolerance test points are normalized according to the corresponding tolerance limit, and the unilateral tolerance test points are normalized according to the corresponding unilateral tolerance upper limit. When the measured deviation value is in the prohibited deviation direction, a penalty mapping process is performed to obtain the normalized deviation value of each test point. The corresponding discrete state symbol is determined based on the normalized deviation value of each detection point, and the discrete state symbol is combined according to the preset detection point order to generate a spatial pattern signature for characterizing the overall deviation shape of the white body door opening. The quality severity index of the corresponding vehicle body is calculated based on the normalized deviation value of each detection point, and the quality event objects, which include at least vehicle identification, measurement time, normalized deviation value, spatial pattern signature and quality severity index, are arranged in the order of measurement time to form a quality digital thread; In the quality digital thread, the frequency of occurrence of target signatures is counted according to the sliding time window, and the abnormal clustering degree is calculated in combination with the quality severity index within the sliding time window; When the abnormal clustering degree exceeds a preset threshold, the corresponding abnormal time period is determined, and the spatial pattern signature and normalized deviation value within the abnormal time period are mapped to virtual semantic anchor points in the digital twin model of the white body door opening that correspond one-to-one with each detection point for visualization playback.

2. The method for visualizing and tracing CMM quality data of body-in-white door openings according to claim 1, characterized in that, The detection points include at least: a measuring point at the position of the hinge on the door opening, a measuring point at the position of the door lock hook, a measuring point at the position of the hinge on the lower part of the door opening, and a measuring point at the position of the door sill. The acquisition of CMM inspection data for multiple inspection points in the white body door opening area includes: The measured deviation value is obtained from the coordinate measuring machine, and the vehicle body markings and production-related information are obtained from the production execution system.

3. The method for visualizing and tracing CMM quality data of body-in-white door openings according to claim 1, characterized in that, The normalization process for the symmetrical tolerance measurement points according to the corresponding tolerance limits specifically includes: calculating the normalized deviation value using the following formula: ， in, This represents the normalized deviation value. Indicates the first The car body is in the first The measured deviation value of each measuring point Indicates the first Symmetrical tolerance limits for each measuring point; The process of normalizing the unilateral tolerance measurement points according to the corresponding unilateral tolerance upper limit, and performing penalty mapping processing when the measured deviation value is in the prohibited deviation direction, to obtain the normalized deviation value of each detection point, specifically includes: If the measured deviation value is in the permissible direction, the normalized deviation value is calculated using the following formula: ， in, Indicates the first The upper limit of the single-sided tolerance for each measuring point, with an allowable deviation range of [0, ...]. ], ≥0 indicates an allowed direction. <0 indicates a prohibited direction; If the measured deviation value is in the prohibited direction, the normalized deviation value is calculated using the penalty mapping function, which is expressed as: ， in, It is represented as the negative deviation extreme value penalty factor, λ>

0.

4. The method for visualizing and tracing CMM quality data of body-in-white door openings according to claim 1, characterized in that, The step of determining the corresponding discrete state symbol based on the normalized deviation value of each detection point, and combining the discrete state symbols according to a preset detection point order to generate a spatial pattern signature for characterizing the overall deviation morphology of the body-in-white doorway includes: The corresponding discrete state symbol is determined based on the normalized deviation value of each detection point, and the spatial pattern signature is formed by combining the symbols according to the preset detection point order. The status level of each detection point is determined based on the absolute value of its normalized deviation, wherein the status level includes at least normal state, warning state, and out-of-tolerance state. A discrete status symbol is determined for each detection point based on a comparison between its normalized deviation and a preset discrete threshold. When the absolute value of the normalized deviation is not greater than the preset discrete threshold, the discrete status symbol is a normal symbol; when the normalized deviation is greater than the preset discrete threshold, the discrete status symbol is a positive deviation symbol; and when the normalized deviation is less than the negative of the preset discrete threshold, the discrete status symbol is a negative deviation symbol. The discrete status symbols corresponding to each detection point are combined according to a preset detection point order to form the spatial pattern signature.

5. The method for visualizing and tracing CMM quality data of body-in-white door openings according to claim 1, characterized in that, The quality severity index is used to characterize the severity of the overall vehicle body quality. The process involves calculating the corresponding vehicle body quality severity index based on the normalized deviation value of each detection point, and arranging quality event objects, including at least vehicle identification, measurement time, normalized deviation value, spatial pattern signature, and quality severity index, in chronological order to form a quality digital thread, including: The quality severity index of the corresponding vehicle body is calculated based on the normalized deviation value of each detection point. The quality severity index is the maximum value among the absolute values ​​of the normalized deviation values ​​corresponding to multiple detection points, and is expressed by the following formula: ； After obtaining the normalized deviation value, spatial pattern signature, and quality severity index, the detection results of each vehicle body are encapsulated into a quality event object. The quality event object includes: vehicle body identification, measurement time, normalized deviation vector, quality severity index, and spatial pattern signature. Multiple quality event objects are arranged in chronological order to form a quality digital thread corresponding to the doorway area of ​​the white body.

6. The method for visualizing and tracing CMM quality data of body-in-white door openings according to claim 1, characterized in that, In the quality digital thread, the frequency of occurrence of target signatures is counted according to a sliding time window, and the anomaly clustering degree is calculated in combination with the quality severity index within the sliding time window, including: In the quality digital thread, time windows are divided according to a preset time span or a preset number of vehicles. Spatial pattern signatures of continuously produced vehicles are statistically analyzed using a sliding time window. Anomaly clustering is calculated by combining the frequency of target signature occurrences and the quality severity index. Specifically, this includes: Let the current sliding time window be W(t,Δt), which contains N vehicle bodies; the number of times a certain target signature appears within this time window is... The abnormal clustering degree is calculated using the following formula. : in, Indicates the number of vehicles. This indicates the frequency of the label name within the current time window. This indicates the severity index of the quality. This indicates the overall severity level of the current time window.

7. The method for visualizing and tracing CMM quality data of body-in-white door openings according to claim 1, characterized in that, When the abnormal clustering degree exceeds a preset threshold, the corresponding abnormal time period is determined, and the spatial pattern signature and normalized deviation value within the abnormal time period are mapped to virtual semantic anchor points in the digital twin model of the white body doorway that correspond one-to-one with each detection point for visualization playback, including: If the abnormal clustering degree exceeds a preset threshold, the corresponding abnormal time period is determined, and the spatial pattern signature and normalized deviation value within the abnormal time period are mapped to the digital twin model of the body-in-white doorway. The virtual semantic anchor points are then visualized and replayed. The virtual semantic anchor points are preset in the digital twin model of the body-in-white doorway and correspond one-to-one with each of the detection points. The visualization and replay includes color mapping display of each virtual semantic anchor point based on the normalized deviation value. Establish a pre-defined association table to characterize the relationship between spatial pattern signatures and process objects; Based on the preset association table, investigation clues are generated.

8. A visual traceability system for CMM quality data of body-in-white door openings using the method described in any one of claims 1-7, characterized in that, include: The data acquisition module is used to acquire CMM inspection data from multiple inspection points in the doorway area of ​​the white body. The CMM inspection data includes at least the body markings, measurement time, and the measured deviation value corresponding to each inspection point. The heterogeneous tolerance normalization module is used to perform heterogeneous tolerance normalization processing on the measured deviation values ​​of the corresponding detection points according to the tolerance type of each detection point, so as to obtain the normalized deviation value of each detection point. The tolerance types of the detection points include: symmetrical tolerance measurement points and unilateral tolerance measurement points. The symmetrical tolerance measurement points are normalized according to the corresponding tolerance limit, and the unilateral tolerance measurement points are normalized according to the corresponding unilateral tolerance upper limit. When the measured deviation value is in the prohibited deviation direction, a penalty mapping process is performed to obtain the normalized deviation value of each detection point. The pattern signature generation module is used to determine the corresponding discrete state symbols based on the normalized deviation values ​​of each detection point, and to combine the discrete state symbols according to the preset detection point order to generate a spatial pattern signature that characterizes the overall deviation shape of the white body door opening. The Quality Digital Thread Construction Module is used to calculate the corresponding vehicle body quality severity index based on the normalized deviation value of each detection point, and to arrange quality event objects, including at least vehicle body markings, measurement time, normalized deviation value, spatial pattern signature and quality severity index, in the order of measurement time to form a quality digital thread; The clustering analysis module is used to count the frequency of occurrence of target signatures according to a sliding time window in the quality digital thread, and to calculate the abnormal clustering degree in combination with the quality severity index within the sliding time window. The visualization traceability module is used to determine the corresponding abnormal time period when the abnormal clustering degree exceeds a preset threshold, and to map the spatial pattern signature and normalized deviation value within the abnormal time period to virtual semantic anchor points in the digital twin model of the white body door opening that correspond one-to-one with each detection point, so as to perform visualization playback and output the investigation clues of the corresponding tooling or fixture.

9. An electronic device, characterized in that, The method includes a processor and a memory, the memory being connected to the processor; the memory is used to store a computer program; the processor invokes the computer program stored in the memory to perform the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a computer, performs the method according to any one of claims 1-7.

Images