A method and system for detecting the sealing performance of an automobile sunroof sealing strip

Abstract

The application provides a kind of automobile sunroof sealing strip sealing performance detection method and system, it is related to performance detection field, the method includes: obtaining automobile sunroof processing log, obtains abnormal calibration result by abnormal calibration, traverses calibration result and deploys wind force sensor, closes the air flow channel inside and outside automobile, activates wind force gun, collects wind force vector set, by sealing abnormality analysis component, carries out feature fusion, obtains sealing abnormality level label, finally, sealing abnormality level label and abnormal position calibration result are associated storage and sent to sealing performance detection management end. Through the present application, the technical problems of existing technology, such as difficult abnormal position positioning, lack of automated detection, difficult traceability and management, and inaccurate performance detection, can be solved. The accuracy and efficiency of automobile sunroof sealing performance detection can be improved, production costs can be reduced, and product quality and customer satisfaction can be improved.

Description

Technical Field

[0001] This application relates to the field of performance testing technology, and in particular to a method and system for testing the sealing performance of automotive sunroof sealing strips. Background Technology

[0002] As an important part of a vehicle, the sunroof not only provides extra light and ventilation to the cabin but also increases driving and passenger comfort. However, if the sunroof's sealing performance is poor, rainwater, dust, and other external pollutants may enter the cabin, potentially affecting the vehicle's structural integrity and the comfort of its occupants.

[0003] In practical applications, sunroof sealing strips may be affected by various factors, such as temperature, humidity, ultraviolet radiation, and mechanical stress. These factors may cause the sealing strips to age, deform, or break, thus affecting their sealing performance. Therefore, testing methods need to fully consider these factors to ensure that the sealing strips maintain a good sealing effect under various environments.

[0004] Currently, traditional sealing performance testing is usually done manually, which has a low degree of automation and results in poor efficiency.

[0005] In summary, existing technologies face challenges in locating abnormal locations, lack automated detection, and make tracing and management difficult, leading to inaccurate performance testing. Summary of the Invention

[0006] The purpose of this application is to provide a method and system for testing the sealing performance of automotive sunroof sealing strips, in order to solve the technical problems of inaccurate performance testing caused by difficulties in locating abnormal positions, lack of automated testing, and difficulties in traceability and management in the prior art.

[0007] In view of the above problems, this application provides a method and system for testing the sealing performance of automotive sunroof sealing strips.

[0008] Firstly, this application provides a method for testing the sealing performance of automotive sunroof sealing strips. The method is implemented using a system for testing the sealing performance of automotive sunroof sealing strips. The method includes: obtaining an automotive sunroof processing log, which includes automotive model information and sunroof sealing process records; performing sealing anomaly calibration based on the automotive model information and the sunroof sealing process records to obtain anomaly location calibration results; deploying wind sensors through the anomaly location calibration results, closing the airflow channels inside and outside the vehicle, activating a wind gun to blow air according to a preset output power and preset blowing position, and collecting wind vector sets from the wind sensors, wherein the anomaly location calibration results and the wind vector sets correspond one-to-one; performing feature fusion on the wind vector sets using a sealing anomaly analysis component to obtain a sealing anomaly level label; associating and storing the sealing anomaly level label and the anomaly location calibration results, and sending them to a sealing performance testing management terminal.

[0009] Secondly, this application also provides a sealing performance testing system for automotive sunroof sealing strips, used to perform a sealing performance testing method for automotive sunroof sealing strips as described in the first aspect, wherein the system includes: a processing log acquisition module, the processing log acquisition module being used to acquire automotive sunroof processing logs, wherein the automotive sunroof processing logs include automotive model information and sunroof sealing process records; an abnormal location acquisition module, the abnormal location acquisition module being used to perform sealing abnormality calibration based on the automotive model information and the sunroof sealing process records, and obtain abnormal location calibration results; and a wind speed vector set acquisition module, the wind speed vector set acquisition module being used to... The system iterates through the abnormal location calibration results, deploys wind sensors, closes the airflow channels inside and outside the vehicle, activates a wind gun to blow air according to a preset output power and preset blowing position, and iterates through the abnormal location calibration results to collect the wind vector set of the wind sensors. The abnormal location calibration results and the wind vector set correspond one-to-one. A feature fusion module is used to perform feature fusion on the wind vector set through a sealing anomaly analysis component to obtain a sealing anomaly level label. An association storage module is used to associate and store the sealing anomaly level label and the abnormal location calibration results, and send them to the sealing performance detection management terminal.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages: By obtaining the automotive sunroof processing log, which includes automotive model information and sunroof sealing process records; performing sealing anomaly calibration based on the automotive model information and the sunroof sealing process records to obtain anomaly location calibration results; deploying wind sensors through the anomaly location calibration results, closing the airflow channels inside and outside the vehicle, activating a wind gun to blow air according to a preset output power and preset blowing position, and collecting wind force vector sets from the wind sensors, wherein the anomaly location calibration results and the wind force vector sets correspond one-to-one; using a sealing anomaly analysis component, performing feature fusion on the wind force vector sets to obtain a sealing anomaly level label; associating and storing the sealing anomaly level label and the anomaly location calibration results, and sending them to the sealing performance testing management terminal, this effectively solves the technical problems of difficulty in locating anomalies, lack of automated detection, and difficulty in traceability and management in existing technologies, leading to inaccurate performance testing. It can improve the accuracy and efficiency of automotive sunroof sealing performance testing, reduce production costs, and improve product quality and customer satisfaction.

[0011] The above description is merely an overview of the technical solution of this application. To better understand the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0013] Figure 1 This is a flowchart illustrating a method for testing the sealing performance of an automotive sunroof sealing strip according to this application. Figure 2 This is a schematic diagram of the sealing performance testing system for an automotive sunroof sealing strip according to this application.

[0014] Explanation of reference numerals in the attached figures: Processing log acquisition module 11, abnormal location acquisition module 12, wind speed vector set acquisition module 13, feature fusion module 14, associated storage module 15. Detailed Implementation

[0015] This application provides a method and system for testing the sealing performance of automotive sunroof sealing strips, which solves the technical problems of inaccurate performance testing caused by difficulties in locating abnormal positions, lack of automated testing, and difficulties in traceability and management in the prior art. It can improve the accuracy and efficiency of automotive sunroof sealing performance testing, reduce production costs, and improve product quality and customer satisfaction.

[0016] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0017] Example 1 Please see the appendix Figure 1 This application provides a method for testing the sealing performance of an automotive sunroof sealing strip. The method is applied to a system for testing the sealing performance of an automotive sunroof sealing strip, and specifically includes the following steps: Step 1: Obtain the vehicle sunroof processing log, which includes vehicle model information and sunroof sealing process records; Specifically, the acquired automotive sunroof processing logs include, but are not limited to, vehicle model information and sunroof sealing process records. Those skilled in the art pre-define the log format, such as data fields, data types, and data units. The storage location for the automotive sunroof processing logs is determined; this could be a database, file server, or cloud storage. A connection is established with the data source based on its type and access permissions. Based on the data source type and log format, corresponding query statements are written to retrieve and extract vehicle model information and sunroof sealing process records. The query statements are run to retrieve and extract the required log data from the data source. The extracted data is validated to ensure its accuracy and completeness. The extracted raw data is cleaned, including removing duplicate data, handling missing values, and correcting erroneous data. The cleaned data is then formatted according to a predefined format. The formatted data is stored in a suitable location, such as a database, data warehouse, or file, for subsequent use and analysis. The vehicle model information includes the model, model year, model number, production batch, production location, and vehicle configuration overview. The sunroof sealing process record includes the type of sealing strip, the specifications of the sealing strip, the auxiliary materials used such as cleaning agents, lubricants, adhesives, and the application trajectory.

[0018] Step 2: Based on the vehicle model information and the sunroof sealing process record, perform sealing anomaly calibration to obtain the anomaly location calibration results; Specifically, the process involves matching vehicle model information with sunroof sealing process records. Database queries ensure that each vehicle model is associated with its corresponding sealing process record. Different sealing process standards may exist for different vehicle models, and this includes converting various parameters such as coating trajectory and flow rate into comparable units or indicators. By analyzing the standardized data, outliers that deviate from the normal process parameter range are identified. Statistical methods such as mean and standard deviation analysis, or machine learning algorithms such as anomaly detection algorithms, are used. Once outliers are identified, their specific locations are determined. For example, the original sunroof sealing process records can be reviewed to find the corresponding coating trajectory. Detailed information for each anomaly location is recorded, including the vehicle model, anomaly type (e.g., insufficient or excessive coating), and a specific description of the anomaly location (e.g., which part of the sunroof, which point on the coating trajectory).

[0019] Step 3: Deploy wind sensors by traversing the abnormal location calibration results, close the airflow channels inside and outside the car, activate the wind gun to blow air according to the preset output power and preset blowing position, traverse the abnormal location calibration results, and collect the wind vector set of the wind sensor, wherein the abnormal location calibration results and the wind vector set correspond one-to-one. Specifically, ensure a sufficient number of calibrated wind sensors capable of accurately measuring wind magnitude and direction. Prepare a wind gun with adjustable output power and blowing position, ensuring it functions correctly and meets testing requirements. Close the airflow channels inside and outside the vehicle to eliminate interference from external airflow. Based on the calibration results of abnormal locations, determine the testing sequence, ensuring each abnormal location is effectively tested. Deploy a wind sensor near each abnormal location, ensuring it accurately measures the wind force generated by the wind gun. Ensure the sensors are securely fixed at the test locations, preventing movement or shaking during blowing. Set the output power and blowing position of the wind gun according to testing requirements. Ensure these parameters remain consistent throughout the test. Perform blowing tests on each abnormal location according to the planned testing sequence. At each location, activate the wind gun and blow according to preset parameters. During blowing, record the wind vector measured by each wind sensor, including magnitude and direction, in real time. Ensure these data correspond one-to-one with the abnormal locations. Organize the collected wind vector data into a complete wind vector set. This set should contain wind vector information for all test locations. Based on the wind vector set, analyze the wind distribution at anomalous locations. This helps to understand the extent to which sealing anomalies affect airflow and potential leakage paths.

[0020] Step 4: Using the sealing anomaly analysis component, perform feature fusion on the wind vector set to obtain the sealing anomaly level label; Specifically, ensure the wind vector set data is complete and accurate, removing any outliers or erroneous data. Standardize the features in the wind vector set, such as wind magnitude and direction, to eliminate the influence of units and dimensions, making different features comparable. Select features related to sealing anomalies from the wind vector set. These features may include wind magnitude, direction, stability, etc. Choose an appropriate feature fusion method based on the nature and importance of the selected features, including weighted average, principal component analysis, and neural network fusion. Using the selected feature fusion method, fuse multiple features into one or more comprehensive features that more comprehensively reflect the sealing anomaly situation. Define different levels of sealing anomalies according to actual needs, such as slight, moderate, and severe. Set corresponding thresholds for each anomaly level based on historical data. These thresholds can be determined by expert experience or statistical methods. Compare the fused comprehensive features with the set thresholds and assign a corresponding sealing anomaly level label to each anomaly location.

[0021] Step 5: Associate and store the sealing anomaly level label and the anomaly location calibration result, and send them to the sealing performance testing management terminal.

[0022] Specifically, the sealed anomaly level labels obtained after feature fusion are also organized in a unified format to ensure that each label corresponds to a specific anomaly location. The number of anomaly location labeling results is checked against the number of sealed anomaly level labels. If any mismatch exists, data adjustments or supplementation are needed to ensure a one-to-one correspondence. A unique identifier (such as a number or UUID) is generated for each item in the anomaly location labeling results, ensuring its uniqueness within the dataset. The generated unique identifiers are then mapped to the sealed anomaly level labels. Based on the data volume and query requirements, a suitable database management system or storage system is selected, such as a relational database like MySQL or a NoSQL database like MongoDB. Table structures are created or updated in the database to store the anomaly location labeling results and sealed anomaly level labels, reflecting the relationships between them. Table design should consider query efficiency and scalability. The organized data is imported into the database, ensuring that the associated key-value pairs are stored correctly and reflect the relationship between the labeling results and the labels. According to the management requirements, scripts are written or existing data transformation tools are used to convert the data in the database into a format acceptable to the management, such as CSV, XML, or JSON. Choose an appropriate transmission protocol based on the sensitivity and security requirements of the data. If the data contains sensitive information, an encrypted transmission protocol, such as SFTP or HTTPS, should be used. Configure necessary transmission parameters according to the selected protocol, such as server address, port number, and authentication method. Write automated scripts to extract data from the source database periodically or on demand, format it into a specified format, and then send it to the target management terminal via the configured transmission protocol. Alternatively, if the data volume is small or changes infrequently, these steps can be performed manually. Configure a corresponding service on the sealing performance testing management terminal to receive the data; this could be an FTP server, web server, or dedicated API interface. The receiving service should authenticate the sender and perform integrity checks on the received data to ensure it has not been tampered with during transmission. The management terminal needs to store the received data in its own database for subsequent querying and analysis, including steps such as data parsing, cleaning, and transformation.

[0023] Furthermore, step two of this application also includes: The sunroof sealing process record includes sealant application trajectory information, sealant application flow rate information, and sealant application duration information; Using the sealant application trajectory information, sealant application flow rate information, sealant application duration information, and vehicle model information as retrieval constraints, a sealant detection record dataset is collected. The sealing detection record dataset is traversed to extract the sealing anomaly detection locations, obtaining the first sealing anomaly location set, the second sealing anomaly location set, up to the Nth sealing anomaly location set, where N≥1 and N is an integer; High-frequency position analysis is performed on the first set of sealing abnormal positions, the second set of sealing abnormal positions, up to the Nth set of sealing abnormal positions, to obtain the abnormal position calibration results.

[0024] Specifically, sunroof sealing process records are collected, including sealant application trajectory information, sealant application flow rate information, and sealant application duration information. Vehicle model information is used as an additional retrieval constraint. Different vehicle models may have different sunroof designs and sealing requirements, therefore model information is necessary for accurate analysis. Using the aforementioned sealant application trajectory, flow rate, duration, and vehicle model information as retrieval constraints, relevant sealing inspection record datasets are retrieved from the database. This dataset will contain multiple historical records of sealing inspections. The retrieved sealing inspection record dataset is traversed, and the content of each record is analyzed. For each record, anomaly detection algorithms or preset anomaly judgment criteria are applied to identify the location of sealing anomalies. These criteria may be based on trajectory deviation, flow rate anomalies, or excessive / insufficient application duration, etc. The identified anomaly locations are categorized and stored in different sets, such as the first sealing anomaly location set, the second sealing anomaly location set, and so on up to the Nth sealing anomaly location set. Here, N represents the number of anomaly location sets, which is an integer greater than or equal to 1. Among them, the sealant includes the brand and model used, the application trajectory includes the path or pattern, and the flow information includes the application rate or quantity.

[0025] Furthermore, this application also includes: Construct a sealing process deviation evaluation function: , , , , in, Characterizing the first sealing process record, Characterizing the i-th position of the first sealing process record, Characterizes the sealant application flow rate at position i in the first sealing process. The duration of sealant application at position i is recorded in the first sealing process. Characterizing the second sealing process record, Characterizing the i-th position of the second sealing process record, Characterizing the sealant application flow rate at position i in the second sealing process, The second sealing process records the sealant application time at position i. Characterizing the first deviation coefficient, The second deviation coefficient is represented by M, the total number of positions recorded in the first sealing process is represented by L, and the total number of positions recorded in the second sealing process is represented by L. The Euclidean distance between the first sealing process record at position i and the second sealing process record at position i, in a unified coordinate system. Characterizing the distance deviation threshold, Characterizing the flow deviation threshold, Characterizing the time deviation threshold, and For pre-configured weights, the sum of the two equals 1. Characterizing the second deviation coefficient; Using the vehicle model information as the primary retrieval constraint, a first set of sealing inspection record datasets is collected; Traverse the first sealing process record set of the first sealing detection record dataset and evaluate the deviation from the sealant application trajectory information, the sealant application flow rate information, and the sealing application duration information to obtain a binary deviation coefficient set; Based on the first deviation coefficient threshold and the second deviation coefficient threshold, the first sealing detection record dataset is sorted according to the binary deviation coefficient set to obtain the sealing detection record dataset, wherein the sorting is achieved when both the first deviation coefficient threshold and the second deviation coefficient threshold are satisfied.

[0026] Specifically, the process of obtaining the deviation of the above sealing process from the evaluation function is as follows: First, let the standard record for the automotive sunroof sealing process be... The actual record to be evaluated is Recorded as: Standard Process Record: Actual process record: .

[0027] Then, construct the position deviation evaluation function (first deviation coefficient): Step A: Define the single-point position deviation. For any i-th position, define the Euclidean distance between the two trajectory positions: , where (x,y,z) are the coordinate values ​​in the coordinate system.

[0028] Step B: Introduce a distance threshold and determine the criteria for judging positional deviation. Set the distance deviation threshold as follows: A positional deviation is defined as occurring when the deviation between two records at the same position exceeds or equals a threshold. .

[0029] Step C: Count the number of deviations for all positions on the trajectory. To ensure that each point corresponds effectively, the number of positions counted is the shorter of the two records, min(M,L). .

[0030] Step D: Construct the position deviation coefficient formula (normalization process) to ensure that the coefficients are comparable and fall within the range of 0.5. The interval is normalized using the larger of the total number of positions, max(M,L), as the denominator: .

[0031] Next, we construct the evaluation function for the deviation between flow rate and duration (the second deviation coefficient): Step A: Define the criteria for determining single-point flow and duration deviation, given the flow deviation threshold Δflow and the duration deviation threshold Δt, and define the flow and duration deviation functions for a single point: Flow deviation: Duration deviation: .

[0032] Step B: Count the number of deviations between flow rate and duration for all locations. Count the flow rate and duration separately, up to the min(M,L)th location: Total number of flow rate deviations: Total number of duration deviations: .

[0033] Step C: Define weighting coefficients to weight and synthesize the deviation between traffic and duration. Since traffic and duration have different importance, weighting coefficients are introduced. ,in =1, defined as: .

[0034] Step D: Construct the flow duration deviation coefficient (normalized), normalize using the larger number of locations max(M,L) as the denominator, and obtain the second deviation coefficient: Wherein, the numerators in the first and second deviation coefficients are adopted. And the denominator adopts This is primarily to allow for a more scientific and fair measurement of process deviations when comparing sealing process records of different lengths. The specific reasons are as follows: Standardize the comparison criteria: The number of locations recorded in the automotive sunroof sealing process may vary, leading to inaccurate direct comparisons of deviations. Use a molecular scale... The denominator is This format can unify the deviations of records of different lengths to the same scale, eliminate the influence of differences in record length, and make the deviation coefficients more comparable.

[0035] Highlighting the effective comparison range: Molecular We only focus on the corresponding positions that appear in both sealing process records. For example, if the two records have 10 and 8 positions respectively, we only compare the deviations of these 8 corresponding positions to avoid interference from additional positions and accurately reflect the degree of process deviation at the same position.

[0036] Control the range of deviation coefficients: use the denominator The deviation coefficient can be controlled within the range of [0, 1]. The coefficient is 1 when the deviation is maximum and 0 when they are exactly the same. This makes it easier to set a threshold to determine whether the process is abnormal. For example, if the first deviation coefficient threshold is set to 0.3, exceeding the threshold may indicate a risk of sealing abnormalities.

[0037] Finally, the sealing process deviation evaluation function is formed by combining these two factors. Through the derivation of the two deviation coefficients mentioned above, the final combination of the sealing process deviation evaluation function is as follows: .

[0038] Characterizing the first sealing process record, Characterizing the i-th position of the first sealing process record, Characterizes the sealant application flow rate at position i in the first sealing process. The duration of sealant application at position i is recorded in the first sealing process. Characterizing the second sealing process record, Characterizing the i-th position of the second sealing process record, Characterizing the sealant application flow rate at position i in the second sealing process, The second sealing process records the sealant application time at position i. Characterizing the first deviation coefficient, The second deviation coefficient is represented by M, the total number of positions recorded in the first sealing process is represented by L, and the total number of positions recorded in the second sealing process is represented by L. The Euclidean distance between the first sealing process record at position i and the second sealing process record at position i, in a unified coordinate system. The distance deviation threshold is characterized by a person skilled in the art (set by the person skilled in the art according to the standard ISO11926-1 "Hydraulic transmission with ISO725 thread and O-ring seals - Ports and stud ends - Part 1: Frustum housing with O-ring seals - Ports, optional, with a sealing surface position deviation ≤0.1mm). The threshold for characterizing flow deviation (set by those skilled in the art according to the standard SAE J200 "Classification System for Rubber Materials", which specifies the fluctuation range of sealant application flow (e.g., ±10%)). The characterization time deviation threshold (to be set by a person skilled in the art according to the standard ASTM D2000 "Classification System for Rubber Products for Automotive Use") It can be based on the vulcanization time calculation formula (such as T90+5%), combined with the production line cycle time set to ±2 seconds. and For pre-configured weights, the sum of the two equals 1. The second deviation coefficient is represented; using vehicle model information as the primary retrieval constraint, a dataset of first-level sealing inspection records related to that model is collected from the database. This dataset should contain multiple sealing process records, each containing information such as the sealant application trajectory, flow rate, and duration. Each sealing process record in the first-level sealing inspection record dataset is traversed and compared with known sealant application trajectory, flow rate, and duration information, and deviation evaluation is performed using a sealing process deviation evaluation function. The purpose of deviation evaluation is to measure the difference between the actual record and the standard or expected value. For each sealing process record, its deviation coefficient from the standard value is calculated. The deviation coefficient can be a numerical value representing the degree of difference between the actual value and the standard value. After calculating the deviation coefficients for all records, a binary deviation coefficient set is obtained, containing deviation coefficients related to trajectory information, flow rate, and duration information.

[0039] The sorting principle for the binary deviation coefficient set is as follows: For each element in the binary deviation coefficient set, check whether its two deviation coefficients simultaneously meet the requirements of the first deviation coefficient threshold and the second deviation coefficient threshold. If both are met (i.e., when De1 ≥ the first threshold and De2 ≥ the second threshold, it is determined to be an abnormal record, where the first and second thresholds are set by those skilled in the art based on a large amount of experimental data), then the record is retained in the new sealing detection record dataset; if not, it is filtered out.

[0040] After filtering, a new and more accurate dataset of sealing inspection records was obtained. This dataset only contains sealing process records that are related to specific car models and meet preset standards.

[0041] Furthermore, this application also includes: Based on the first set of sealing anomaly locations, the second set of sealing anomaly locations, and up to the Nth set of sealing anomaly locations, statistics are performed to obtain the set of anomaly detection locations and the first set of anomaly detection frequencies. Obtain the position deviation threshold; Based on the location deviation threshold, traverse the set of abnormal detection locations and count the frequency set of abnormal detections in neighboring locations; The first anomaly detection frequency set is obtained by summing the neighboring location anomaly detection frequency set that corresponds one-to-one with the first anomaly detection frequency set. Based on the anomaly detection frequency threshold, the anomaly detection location set is sorted according to the second anomaly detection frequency set to obtain the anomaly location calibration result.

[0042] Specifically, statistics are performed on the first set of sealing anomaly locations, the second set of sealing anomaly locations, and so on up to the Nth set of sealing anomaly locations. The purpose of this statistics is to identify all locations where anomalies occur and record the frequency of anomalies at each location. The anomaly locations in all sealing anomaly location sets are merged to form a set containing all unique anomaly locations. For each location in the anomaly detection location set, the total number of times it appears in all sealing anomaly location sets is calculated, forming the first anomaly detection frequency set. Each element in this set represents the anomaly detection frequency at a specific location. The selection of the location deviation threshold should be determined based on the specific sealing process and application scenario; factors that may need to be considered include the application accuracy of the sealant, the size and shape of the skylight, etc. Using the obtained location deviation threshold, the anomaly detection location set is traversed. For each location, its neighboring locations (i.e., other locations with a distance less than or equal to the location deviation threshold) are found, and the anomaly detection frequency of these neighboring locations is statistically analyzed, forming the neighboring location anomaly detection frequency set. The purpose of this step is to account for the possibility that anomalies may not only occur at the same precise location but also at adjacent locations. The obtained neighboring location anomaly detection frequency set is summed with the obtained first anomaly detection frequency set. The anomaly detection frequency of a given location is added to the anomaly detection frequencies of its neighboring locations to obtain a second set of anomaly detection frequencies. A more comprehensive anomaly assessment is obtained by comprehensively considering the anomaly frequencies of each location and its neighboring locations. An anomaly detection frequency threshold is set by those skilled in the art. This threshold is used to determine which locations have anomaly frequencies high enough to be considered true anomalies. Then, based on the second set of anomaly detection frequencies and this threshold, the set of anomaly detection locations is sorted to obtain the anomaly location labeling results.

[0043] Furthermore, step four of this application includes: The interactive wind gun control component obtains the blowing vector set of the wind vector set; Calculate the set of vector cosine eigenvalues ​​based on the wind force vector set and the wind blowing vector set; Calculate the set of shrinkage percentage feature values ​​based on the wind force vector set and the wind blowing vector set; Based on the vector cosine eigenvalue set and the reduction percentage eigenvalue set, construct the vector cosine feature matrix and the reduction percentage feature matrix; The sealing anomaly analysis component analyzes the vector cosine feature matrix and the reduction percentage feature matrix to obtain the sealing anomaly level label.

[0044] Specifically, wind vector data is acquired through the control components of a wind gun. The wind gun may generate a series of wind vectors, representing the direction and force of the wind at different times and settings. This data is collected to form a set of wind vectors, where each vector represents the wind conditions at a specific point in time. The wind vector set and the wind vector set are used to calculate the cosine value between the vectors. The cosine value is a number between -1 and 1, used to measure the similarity between two vectors. A cosine value of 1 indicates that the two vectors are identical, -1 indicates they are completely opposite, and 0 indicates they are orthogonal (i.e., unrelated). The cosine values ​​between all wind vectors and wind vectors are calculated, forming a set of vector cosine eigenvalues. A reduced percentage eigenvalue set is then calculated. This eigenvalue set may involve comparing the difference between the magnitude of the wind vector and the magnitude of the wind vector and converting it into a percentage form. Using the vector cosine eigenvalue set and the reduced percentage eigenvalue set, two feature matrices are constructed: a vector cosine feature matrix and a reduced percentage feature matrix. These matrices organize the eigenvalues ​​in a two-dimensional table for easier subsequent analysis and processing. The two feature matrices mentioned above are analyzed using a sealing anomaly analysis component. This analysis process may include pattern recognition, statistical analysis, or machine learning methods to identify anomalous behaviors that deviate from normal operating patterns. Based on these analysis results, a sealing anomaly level label is assigned to each wind vector or a series of vectors. This label may be a number or a category such as "normal," "minor anomaly," or "major anomaly," indicating potential problems during the sealing process.

[0045] Furthermore, this application also includes: Calculate the set of absolute values ​​of the magnitude deviation between the wind force vector set and the blowing vector set; Calculate the percentage of the absolute value of the modulus deviation in the set of blowing vectors to obtain the set of feature values ​​of the reduction percentage.

[0046] Specifically, for each vector in the wind vector set and the blowing vector set, its magnitude is calculated. For each corresponding wind vector and blowing vector, we need to calculate the deviation of their magnitudes. The deviation can be calculated using the following formula: Deviation = |Wind Vector Magnitude - Blowing Vector Magnitude|. The absolute values ​​of all calculated magnitude deviations are collected to form a set of absolute magnitude deviations. For each value in the set of absolute magnitude deviations, its percentage in the corresponding blowing vector magnitude is calculated. The percentage can be calculated using the following formula: Percentage = Absolute Value of Blowing Vector Magnitude Deviation × 100%. The percentages are collected to form a set of reduction percentage feature values. Each value in this set represents the percentage reduction in magnitude for the corresponding wind vector and blowing vector.

[0047] Furthermore, this application also includes: Collect a sealing anomaly detection record dataset, wherein the sealing anomaly detection record dataset includes wind force vector record data, wind blowing vector record data, and sealing anomaly level ground truth; Based on the wind vector recording data and the wind force vector recording data, construct a vector cosine recording matrix, a reduced percentage recording matrix, and the true value of the sealing anomaly level; The sealing anomaly analysis component is configured using the true value of the sealing anomaly level as supervisory data and the vector cosine recording matrix and the reduction percentage recording matrix as inputs.

[0048] Specifically, data is collected from the actual sealing process. This dataset should include wind vector log data, blowing vector log data, and corresponding ground truth values ​​for sealing anomaly levels. This data can be collected in real time during the sealing operation using various sensors and monitoring systems. Wind vector log data refers to the recordings of the direction and intensity of wind generated by a wind gun at different times and settings. Blowing vector log data refers to the recordings of the direction and intensity of wind actually acting on the sealing material. The sealing anomaly level ground truth values, determined by a quality inspection system, identify whether a sealing anomaly occurred under specific wind and blowing conditions, and the severity of the anomaly. The necessary matrices are constructed. A vector cosine log matrix is ​​created by calculating the cosine similarity between each wind vector log and its corresponding blowing vector log, and organizing these values ​​into a matrix. A reduced percentage log matrix is ​​created by calculating the percentage difference between the modulus of each wind vector log and its corresponding blowing vector log, and organizing these values ​​into another matrix. This matrix will reveal the attenuation of wind force as it is transmitted to the sealing material. The sealing anomaly analysis component is configured, and the working principle of the model is preferably to construct a two-dimensional coordinate system with vector cosine and reduction percentage as coordinate axes. Then, the user divides the sealing anomaly level region into the two-dimensional coordinate system. Generally, the larger the angle and the larger the reduction percentage, the higher the level. Then, several levels are obtained based on the output matrix set. The model will fit the mode values ​​of several levels as the output value of the model.

[0049] In summary, the method for testing the sealing performance of an automotive sunroof sealing strip provided in this application has the following technical advantages: By obtaining the automotive sunroof processing log, which includes automotive model information and sunroof sealing process records; performing sealing anomaly calibration based on the automotive model information and the sunroof sealing process records to obtain anomaly location calibration results; deploying wind sensors through the anomaly location calibration results, closing the airflow channels inside and outside the vehicle, activating a wind gun to blow air according to a preset output power and preset blowing position, and collecting wind force vector sets from the wind sensors, wherein the anomaly location calibration results and the wind force vector sets correspond one-to-one; using a sealing anomaly analysis component, performing feature fusion on the wind force vector sets to obtain a sealing anomaly level label; associating and storing the sealing anomaly level label and the anomaly location calibration results, and sending them to the sealing performance testing management terminal, this effectively solves the technical problems of difficulty in locating anomalies, lack of automated detection, and difficulty in traceability and management in existing technologies, leading to inaccurate performance testing. It can improve the accuracy and efficiency of automotive sunroof sealing performance testing, reduce production costs, and improve product quality and customer satisfaction.

[0050] Example 2 Based on the sealing performance testing method for an automotive sunroof sealing strip described in the foregoing embodiments, and using the same inventive concept, this application also provides a sealing performance testing system for an automotive sunroof sealing strip. Please refer to the appendix. Figure 2 The system includes: The processing log acquisition module 11 is used to obtain the processing log of the car sunroof, wherein the processing log of the car sunroof includes car model information and sunroof sealing process records; Anomaly location acquisition module 12 is used to perform sealing anomaly calibration based on the vehicle model information and the sunroof sealing process record, and obtain anomaly location calibration result. The wind speed vector set acquisition module 13 is used to traverse the abnormal location calibration results to deploy wind sensors, close the airflow channels inside and outside the car, activate the wind gun to blow according to the preset output power and preset blowing position, traverse the abnormal location calibration results to collect the wind vector set of the wind sensor, wherein the abnormal location calibration results and the wind vector set correspond one-to-one. Feature fusion module 14 is used to perform feature fusion on the wind vector set through the sealing anomaly analysis component to obtain a sealing anomaly level label; The associated storage module 15 is used to associate and store the sealing anomaly level tag and the anomaly location calibration result, and send them to the sealing performance detection management terminal.

[0051] Furthermore, the abnormal location acquisition module 12 in the system is also used for: The sunroof sealing process record includes sealant application trajectory information, sealant application flow rate information, and sealant application duration information; Using the sealant application trajectory information, sealant application flow rate information, sealant application duration information, and vehicle model information as retrieval constraints, a sealant detection record dataset is collected. The sealing detection record dataset is traversed to extract the sealing anomaly detection locations, obtaining the first sealing anomaly location set, the second sealing anomaly location set, up to the Nth sealing anomaly location set, where N≥1 and N is an integer; High-frequency position analysis is performed on the first set of sealing abnormal positions, the second set of sealing abnormal positions, up to the Nth set of sealing abnormal positions, to obtain the abnormal position calibration results.

[0052] Furthermore, the system also includes a detection record set acquisition module, which is used for: Construct a sealing process deviation evaluation function: , , , , in, Characterizing the first sealing process record, Characterizing the i-th position of the first sealing process record, Characterizes the sealant application flow rate at position i in the first sealing process. The duration of sealant application at position i is recorded in the first sealing process. Characterizing the second sealing process record, Characterizing the i-th position of the second sealing process record, Characterizing the sealant application flow rate at position i in the second sealing process, The second sealing process records the sealant application time at position i. Characterizing the first deviation coefficient, The second deviation coefficient is represented by M, the total number of positions recorded in the first sealing process is represented by L, and the total number of positions recorded in the second sealing process is represented by L. The Euclidean distance between the first sealing process record at position i and the second sealing process record at position i, in a unified coordinate system. Characterizing the distance deviation threshold, Characterizing the flow deviation threshold, Characterizing the time deviation threshold, and For pre-configured weights, the sum of the two equals 1. Characterizing the second deviation coefficient; Using the vehicle model information as the primary retrieval constraint, a first set of sealing inspection record datasets is collected; Traverse the first sealing process record set of the first sealing detection record dataset and evaluate the deviation from the sealant application trajectory information, the sealant application flow rate information, and the sealing application duration information to obtain a binary deviation coefficient set; Based on the first deviation coefficient threshold and the second deviation coefficient threshold, the first sealing detection record dataset is sorted according to the binary deviation coefficient set to obtain the sealing detection record dataset, wherein the sorting is achieved when both the first deviation coefficient threshold and the second deviation coefficient threshold are satisfied.

[0053] Furthermore, the system also includes a calibration result acquisition module, which is used for: Based on the first set of sealing anomaly locations, the second set of sealing anomaly locations, and up to the Nth set of sealing anomaly locations, statistics are performed to obtain the set of anomaly detection locations and the first set of anomaly detection frequencies. Obtain the position deviation threshold; Based on the location deviation threshold, traverse the set of abnormal detection locations and count the frequency set of abnormal detections in neighboring locations; The first anomaly detection frequency set is obtained by summing the neighboring location anomaly detection frequency set that corresponds one-to-one with the first anomaly detection frequency set. Based on the anomaly detection frequency threshold, the anomaly detection location set is sorted according to the second anomaly detection frequency set to obtain the anomaly location calibration result.

[0054] Furthermore, the feature fusion module 14 in the system is also used for: The interactive wind gun control component obtains the blowing vector set of the wind vector set; Calculate the set of vector cosine eigenvalues ​​based on the wind force vector set and the wind blowing vector set; Calculate the set of shrinkage percentage feature values ​​based on the wind force vector set and the wind blowing vector set; Based on the vector cosine eigenvalue set and the reduction percentage eigenvalue set, construct the vector cosine feature matrix and the reduction percentage feature matrix; The sealing anomaly analysis component analyzes the vector cosine feature matrix and the reduction percentage feature matrix to obtain the sealing anomaly level label.

[0055] Furthermore, the system also includes a feature value set acquisition module, which is used for: Calculate the set of absolute values ​​of the magnitude deviation between the wind force vector set and the blowing vector set; Calculate the percentage of the absolute value of the modulus deviation in the set of blowing vectors to obtain the set of feature values ​​of the reduction percentage.

[0056] Furthermore, the analysis component configuration module is used for: Collect a sealing anomaly detection record dataset, wherein the sealing anomaly detection record dataset includes wind force vector record data, wind blowing vector record data, and sealing anomaly level ground truth; Based on the wind vector recording data and the wind force vector recording data, construct a vector cosine recording matrix, a reduced percentage recording matrix, and the true value of the sealing anomaly level; The sealing anomaly analysis component is configured using the true value of the sealing anomaly level as supervisory data and the vector cosine recording matrix and the reduction percentage recording matrix as inputs.

[0057] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Figure 1 The method and specific examples for testing the sealing performance of a car sunroof sealing strip in Embodiment 1 are also applicable to the system for testing the sealing performance of a car sunroof sealing strip in this embodiment. Through the foregoing detailed description of the method for testing the sealing performance of a car sunroof sealing strip, those skilled in the art can clearly understand the system for testing the sealing performance of a car sunroof sealing strip in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here. As the system disclosed in the embodiments corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant details can be found in the method section.

[0058] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0059] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for testing the sealing performance of a car sunroof sealing strip, characterized in that, include: Obtain the vehicle sunroof processing log, which includes vehicle model information and sunroof sealing process records; Based on the vehicle model information and the sunroof sealing process record, a sealing anomaly calibration is performed to obtain the anomaly location calibration result. The wind sensor is deployed by traversing the abnormal location calibration results, the airflow channels inside and outside the car are closed, the wind gun is activated and blows air according to the preset output power and preset blowing position, and the wind vector set of the wind sensor is collected. The abnormal location calibration results and the wind vector set are in one-to-one correspondence. By using the sealing anomaly analysis component, feature fusion is performed on the wind vector set to obtain a sealing anomaly level label; The sealing anomaly level label and the anomaly location calibration result are associated and stored, and then sent to the sealing performance testing management terminal. The sunroof sealing process record includes sealant application trajectory information, sealant application flow rate information, and sealant application duration information. Using the sealant application trajectory information, sealant application flow rate information, sealant application duration information, and vehicle model information as retrieval constraints, a sealant detection record dataset is collected. The sealing detection record dataset is traversed to extract the sealing anomaly detection locations, obtaining the first sealing anomaly location set, the second sealing anomaly location set, up to the Nth sealing anomaly location set, where N≥1 and N is an integer; High-frequency position analysis is performed on the first set of sealing abnormality locations, the second set of sealing abnormality locations, and up to the Nth set of sealing abnormality locations to obtain the abnormality location calibration results. Among them, the sealing process deviation evaluation function is constructed as follows: ， ， ， ， in, Characterizing the first sealing process record, Characterizing the i-th position of the first sealing process record, Characterizes the sealant application flow rate at position i in the first sealing process. The duration of sealant application at position i is recorded in the first sealing process. Characterizing the second sealing process record, Characterizing the i-th position of the second sealing process record, Characterizing the sealant application flow rate at position i in the second sealing process, The second sealing process records the sealant application time at position i. Characterizing the first deviation coefficient, The second deviation coefficient is represented by M, the total number of positions recorded in the first sealing process is represented by L, and the total number of positions recorded in the second sealing process is represented by L. The Euclidean distance between the first sealing process record at position i and the second sealing process record at position i, in a unified coordinate system. Characterizing the distance deviation threshold, Characterizing the flow deviation threshold, Characterizing the time deviation threshold, and For pre-configured weights, the sum of the two equals 1. Characterizing the second deviation coefficient; Using the vehicle model information as the primary retrieval constraint, a first set of sealing inspection record datasets is collected; Traverse the first sealing process record set of the first sealing detection record dataset and evaluate the deviation from the sealant application trajectory information, the sealant application flow rate information, and the sealing application duration information to obtain a binary deviation coefficient set; Based on the first deviation coefficient threshold and the second deviation coefficient threshold, the first sealing detection record dataset is sorted according to the binary deviation coefficient set to obtain the sealing detection record dataset, wherein the first deviation coefficient threshold and the second deviation coefficient threshold are simultaneously satisfied for sorting. Among them, the interactive wind gun control component obtains the blowing vector set of the wind vector set; Calculate the set of vector cosine eigenvalues ​​based on the wind force vector set and the wind blowing vector set; Calculate the set of shrinkage percentage feature values ​​based on the wind force vector set and the wind blowing vector set; Based on the vector cosine eigenvalue set and the reduction percentage eigenvalue set, construct the vector cosine feature matrix and the reduction percentage feature matrix; The sealing anomaly analysis component analyzes the vector cosine feature matrix and the reduction percentage feature matrix to obtain the sealing anomaly level label.

2. The method as described in claim 1, characterized in that, High-frequency position analysis is performed on the first set of sealing anomaly locations, the second set of sealing anomaly locations, up to the Nth set of sealing anomaly locations, to obtain the anomaly location calibration results, including: Based on the first set of sealing anomaly locations, the second set of sealing anomaly locations, and up to the Nth set of sealing anomaly locations, statistics are performed to obtain the set of anomaly detection locations and the first set of anomaly detection frequencies. Obtain the position deviation threshold; Based on the location deviation threshold, traverse the set of abnormal detection locations and count the frequency set of abnormal detections in neighboring locations; The first anomaly detection frequency set is obtained by summing the neighboring location anomaly detection frequency set that corresponds one-to-one with the first anomaly detection frequency set. Based on the anomaly detection frequency threshold, the anomaly detection location set is sorted according to the second anomaly detection frequency set to obtain the anomaly location calibration result.

3. The method as described in claim 1, characterized in that, Based on the wind force vector set and the blowing wind vector set, calculate the set of reduction percentage feature values, including: Calculate the set of absolute values ​​of the magnitude deviation between the wind force vector set and the blowing vector set; Calculate the percentage of the absolute value of the modulus deviation in the set of blowing vectors to obtain the set of feature values ​​of the reduction percentage.

4. The method as described in claim 1, characterized in that, The sealing anomaly analysis component analyzes the vector cosine feature matrix and the reduction percentage feature matrix to obtain the sealing anomaly level label, including the following: Collect a sealing anomaly detection record dataset, wherein the sealing anomaly detection record dataset includes wind force vector record data, wind blowing vector record data, and sealing anomaly level ground truth; Based on the wind vector recording data and the wind force vector recording data, construct a vector cosine recording matrix, a reduced percentage recording matrix, and the true value of the sealing anomaly level; The sealing anomaly analysis component is configured using the true value of the sealing anomaly level as supervisory data and the vector cosine recording matrix and the reduction percentage recording matrix as inputs.

5. A sealing performance testing system for automotive sunroof sealing strips, characterized in that, The system comprises: steps for implementing the method according to any one of claims 1 to 4; A processing log acquisition module is used to obtain the processing log of the car sunroof, wherein the processing log of the car sunroof includes car model information and sunroof sealing process records; An abnormal location acquisition module is used to perform sealing anomaly calibration based on the vehicle model information and the sunroof sealing process record, and obtain the abnormal location calibration result. The wind speed vector set acquisition module is used to traverse the abnormal location calibration results, deploy wind sensors, close the airflow channels inside and outside the car, activate the wind gun to blow according to the preset output power and preset blowing position, traverse the abnormal location calibration results, and acquire the wind vector set of the wind sensor, wherein the abnormal location calibration results and the wind vector set correspond one-to-one. The feature fusion module is used to perform feature fusion on the wind vector set through the sealing anomaly analysis component to obtain a sealing anomaly level label; An associated storage module is used to associate and store the sealing anomaly level tag and the anomaly location calibration result, and send them to the sealing performance testing management terminal.

Images