Dynamic quality monitoring system for intelligent assembly detection of automobile door lock

Abstract

The invention provides a dynamic quality monitoring system for intelligent assembly detection of an automobile door lock, and relates to the technical field of quality monitoring, and the system comprises an intelligent touch glove which is used for recording a worker assembly data set; the multi-dimensional collaborative detection module is used for executing anomaly analysis and establishing a multi-dimensional linkage data set; the general detection module is used for carrying out general detection on the automobile door lock and establishing a detection data set; and the collaborative analysis module is used for calling a collaborative analysis network to perform assembly quality anomaly recognition based on the detection data set, the multi-dimensional linkage data set and the worker assembly data set and establish an anomaly recognition result, and the collaborative analysis network comprises a batch collaborative analysis layer and a deep collaborative analysis layer. According to the invention, the technical targets of multi-dimensional real-time monitoring, intelligent analysis and accurate abnormity identification of the automobile door lock assembling process can be realized, and the technical effects of improving the assembling quality, improving the production efficiency, reducing the reworking cost and enhancing the intelligent level of the manufacturing process are achieved.

Description

Technical Field

[0001] This application relates to the technical field of quality monitoring, and particularly to a dynamic quality monitoring system for intelligent assembly detection of automotive door locks. Background Art

[0002] The automotive manufacturing industry is developing towards high automation and intelligence. As a key component, the assembly quality of automotive door locks directly affects the safety of the whole vehicle and the user experience.

[0003] Currently, in the existing automotive door lock assembly detection technologies, there is a lack of real-time monitoring ability for the assembly process, and it is impossible to effectively capture the detailed operation actions of workers, such as pressing force, operation angle, and movement trajectory. Due to individual differences in the manual assembly process, even the same worker may not perform exactly the same actions when assembling at different times, and it is difficult for traditional detection means to identify quality problems caused by different assembly methods. Secondly, the data analysis ability is limited. Existing detection methods are mostly based on the detection results of a single dimension, and fail to integrate multiple sensor data for comprehensive analysis, resulting in low accuracy and coverage of anomaly detection. In addition, the feedback mechanism of the existing technology is relatively lagged. Usually, quality traceability is only carried out after assembly defects are found, and there is a lack of active warning and intelligent optimization means, resulting in delays in the discovery and improvement of quality problems, and affecting the overall production efficiency.

[0004] In summary, there are technical problems in the existing technology that due to the lack of real-time monitoring of the assembly process, limited data analysis ability, and lagged feedback mechanism, it is difficult to timely discover and accurately identify assembly quality problems, further affecting production efficiency, quality stability, and manufacturing cost control. Summary of the Invention

[0005] The purpose of this application is to provide a dynamic quality monitoring system for intelligent assembly detection of automotive door locks, so as to solve the technical problems in the existing technology that due to the lack of real-time monitoring of the assembly process, limited data analysis ability, and lagged feedback mechanism, it is difficult to timely discover and accurately identify assembly quality problems, further affecting production efficiency, quality stability, and manufacturing cost control.

[0006] In view of the above problems, the present application provides a dynamic quality monitoring system for intelligent assembly detection of automotive door locks, including: an intelligent tactile glove, which is used to record a worker assembly data set after signal interaction with the automotive door lock. The worker assembly data set is marked with an automotive door lock number identifier, and the intelligent tactile glove is integrated with a flexible pressure sensor and an inertial measurement unit; a multi-dimensional collaborative detection module, which is used to perform abnormality analysis after receiving the worker assembly data set, configure linkage collection concerns according to the abnormality analysis results, and perform assembly linkage collection based on the linkage collection concerns after detecting the corresponding automotive door lock number, and establish a multi-dimensional linkage data set; a general detection module, which is used to perform general detection on the automotive door lock and establish a detection data set; a collaborative analysis module, which is used to call a collaborative analysis network to perform assembly quality abnormality identification based on the detection data set, the multi-dimensional linkage data set, and the worker assembly data set, and establish an abnormality identification result. The collaborative analysis network includes a batch collaborative analysis layer and a deep collaborative analysis layer.

[0007] The technical solution provided in the present application has at least the following technical effects or advantages: By achieving the technical goals of multi-dimensional real-time monitoring, intelligent analysis, and accurate abnormality identification of the automotive door lock assembly process, the technical effects of improving the assembly quality, increasing the production efficiency, reducing the rework cost, and enhancing the intelligent level of the manufacturing process are achieved.

[0008] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features, and advantages of the present application more obvious and understandable, the following specifically illustrates the specific implementation manners of the present application. It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present application, nor is it used to limit the scope of the present application. Other features of the present application will become easily understandable through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only exemplary, and for those of ordinary skill in the art, other drawings can be obtained based on the provided drawings without creative efforts.

[0010] Figure 1 It is a schematic structural diagram of the dynamic quality monitoring system for intelligent assembly detection of automotive door locks in the present application.

[0011] Figure 2 It is a schematic structural diagram of the collaborative analysis module in the dynamic quality monitoring system for intelligent assembly detection of automotive door locks in the present application.

[0012] Description of the attached drawing reference numerals: Intelligent tactile glove 10, multi-dimensional collaborative detection module 20, general detection module 30, collaborative analysis module 40, batch analysis sub-module 41, in-depth analysis sub-module 42, output module 43. Detailed implementation manners

[0013] By providing a dynamic quality monitoring system for intelligent assembly detection of automotive door locks, the present application solves the technical problems in the prior art that due to the lack of real-time monitoring of the assembly process, limited data analysis capabilities, and a lagging feedback mechanism, it is difficult to timely discover and accurately identify assembly quality problems, further affecting production efficiency, quality stability, and manufacturing cost control. The technical goal of realizing multi-dimensional real-time monitoring, intelligent analysis, and accurate anomaly identification of the automotive door lock assembly process is achieved, and the technical effects of improving assembly quality, increasing production efficiency, reducing rework costs, and enhancing the intelligent level of the manufacturing process are achieved.

[0014] Next, the technical solutions in the present application will be clearly and completely described with reference to the attached drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application. Additionally, it should be noted that for the sake of description, only the parts related to the present application are shown in the attached drawings rather than all of them.

[0015] Please refer to the attached Figure 1 , the present application provides a dynamic quality monitoring system for intelligent assembly detection of automotive door locks, specifically including: An intelligent tactile glove 10, configured to record a worker assembly data set after signal interaction with an automotive door lock. The worker assembly data set is marked with an automotive door lock number identifier, and the intelligent tactile glove 10 is integrated with a flexible pressure sensor and an inertial measurement unit.

[0016] Specifically, the intelligent tactile glove 10 is a wearable device that can sense the external environment and transmit information to the dynamic quality monitoring system for intelligent assembly detection of automotive door locks. It can not only provide hand protection functions but also collect information such as hand movements and contact pressures. After signal interaction with the automotive door lock, it means that the sensing system inside the glove will sense and record various data when a person operates the door lock, such as pressing force, contact time, rotation angle, etc. Then, all relevant data is collected and stored as a data set, that is, the worker assembly data set. The worker assembly data set is marked with an automotive door lock number identifier to indicate which automotive door lock this data is associated with, so that the assembly situation of each door lock can be accurately traced.

[0017] The intelligent tactile glove 10 is integrated with flexible pressure sensors and inertial measurement units. The flexible pressure sensors are electronic components capable of detecting the magnitude of pressure, which can be bent and adapted to the curved surface of the hand, and can measure the pressing force when a person operates the door lock. The inertial measurement unit is a device that combines an accelerometer and a gyroscope, and can measure the motion state of the glove in real time, including speed, direction, and angle changes.

[0018] The multi-dimensional collaborative detection module 20 is used to perform anomaly analysis after receiving the worker assembly dataset, configure linkage collection attention according to the anomaly analysis results, and perform assembly linkage collection based on the linkage collection attention after detecting the corresponding car door lock number, and establish a multi-dimensional linkage dataset.

[0019] Specifically, the multi-dimensional collaborative detection module 20 is used to receive data from multiple sources and perform comprehensive analysis. By receiving the data generated by workers during the assembly process, calculating the anomaly degree of various indicators during the assembly process, and judging whether a certain operation deviates from the standard range, the anomaly analysis result is obtained. Among them, the difference between the current operation and the standard data is compared, and the deviation degree value is calculated to facilitate taking further measures for optimization.

[0020] According to the high or low anomaly degree in the anomaly analysis result, determine the data that needs further attention, and obtain the configuration result of the linkage collection attention. Among them, linkage collection means not only analyzing a certain data alone, but also comprehensively checking multiple related data. For example, if the pressure data of a certain worker is detected to be abnormal, not only the pressure will be concerned, but also the hand posture, fingertip trajectory, etc. of this worker will be analyzed synchronously to find the real cause of the problem. Configuring attention means dynamically adjusting the focus of attention. For example, if it is found that a certain worker's hand angle is abnormal multiple times when tightening the screws, the monitoring frequency of the worker's hand posture is increased, while the monitoring of pressure may remain at a normal level to ensure accurate analysis of key problems.

[0021] Detect the corresponding car door lock number according to the configuration result of the linkage collection attention, perform linkage collection, integrate the collected multiple data, and form a database containing multiple information dimensions, so that subsequent intelligent algorithms can use these data for in-depth learning and optimization.

[0022] The general detection module 30 is used to perform general detection on the car door lock and establish a detection dataset.

[0023] Specifically, the general detection module 30 is applicable to all car door locks, rather than for specific models or specific assembly situations. The general detection module 30 is used to perform comprehensive detection on the car door lock, such as checking whether the car door lock meets the standard requirements, including dimensions, functions, durability, etc., to ensure the quality of the car door lock and establish a detection dataset.

[0024] A collaborative analysis module 40 is used to call a collaborative analysis network to perform assembly quality anomaly identification based on the detection data set, the multi-dimensional linkage data set, and the worker assembly data set, and establish an anomaly identification result. The collaborative analysis network includes a batch collaborative analysis layer and a deep collaborative analysis layer.

[0025] Specifically, the collaborative analysis module 40 uses multiple data sources to perform anomaly identification on the assembly quality to ensure the stability and consistency of the product quality. The collaborative analysis module 40 calls the collaborative analysis network, comprehensively analyzes different types of data, and finally establishes an anomaly identification result to discover and solve possible problems in the assembly process.

[0026] Calling the collaborative analysis network means automatically or manually triggering the collaborative analysis function to analyze the data collected in the current production process. The collaborative analysis network is a computational framework that combines multiple analysis methods, can integrate data from multiple sources, and perform comprehensive analysis. For example, on a production line, the operation data of workers, the measurement data of detection equipment, and the operation data of the automation system will all be collected and input into the collaborative analysis network for joint analysis.

[0027] Next, the collaborative analysis network needs to perform assembly quality anomaly identification based on the detection data set, the multi-dimensional linkage data set, and the worker assembly data set, and find situations that deviate significantly from the normal assembly process. For example, if during the assembly process of a batch of door locks, the pressing force used by the worker is significantly less than the standard value, and at the same time the detection equipment finds that the sealing performance of the door lock is poor, then this situation may be determined as an assembly anomaly.

[0028] The collaborative analysis network internally includes a batch collaborative analysis layer and a deep collaborative analysis layer. The role of the batch collaborative analysis layer is to perform comparative analysis on products of the same batch. The role of the deep collaborative analysis layer is to more deeply analyze the root cause of the anomaly, and perform correlation analysis by combining the target detection data, the multi-dimensional linkage data set, and the worker assembly data set.

[0029] The dynamic quality monitoring system for intelligent assembly detection of automotive door locks can achieve the technical goals of multi-dimensional real-time monitoring, intelligent analysis, and accurate anomaly identification of the automotive door lock assembly process, and achieve the technical effects of improving the assembly quality, increasing the production efficiency, reducing the rework cost, and enhancing the intelligent level of the manufacturing process.

[0030] Furthermore, the intelligent tactile glove 10 is used for: after performing signal interaction with the car door lock, initiating monitoring; activating the flexible pressure sensors at the fingertips and the palm of the intelligent tactile glove 10 to establish a pressure data set; activating the inertial measurement unit to establish an inertial data set, reconstructing the spatial trajectory based on the action feature extraction channel, resolving the hand pose and performing fingertip motion modeling; performing standard action pose matching according to the hand pose and fingertip motion modeling to establish pose anomalies; using the hand pose and fingertip motion modeling to identify collaborative assembly anomalies in the pressure data set to establish assembly anomalies; constructing a worker assembly data set according to the pose anomalies and the assembly anomalies.

[0031] Specifically, after the intelligent tactile glove 10 performs signal interaction with the car door lock and initiates monitoring, it means that when a person's hand touches the door lock and the contact meets the preset activation conditions, the monitoring function is automatically activated. For example, when a finger touches the door lock handle and maintains a certain pressure for more than one second, data recording is started, while a short contact will not trigger the monitoring.

[0032] Activate the flexible pressure sensors at the fingertips and the palm of the intelligent tactile glove 10 to establish a pressure data set, so as to analyze whether the force applied by the person meets the assembly requirements in the follow-up.

[0033] The inertial measurement unit consists of an accelerometer and a gyroscope, and can detect the motion trajectory, angular velocity and acceleration of the hand in space. Activate the inertial measurement unit to establish an inertial data set, which is used to analyze whether a person's operation is stable and whether there are abnormal shakes or jitters.

[0034] Calculate the motion path of the hand in three-dimensional space based on the inertial data set, resolve the hand pose and perform fingertip motion modeling, that is, analyze the rotation angle and position of the hand, and then perform data analysis on information such as the motion trajectory, speed, and force applied by the fingers, and establish a mathematical model to determine whether the worker's operation meets the standards.

[0035] Perform standard action pose matching according to the hand pose and fingertip motion modeling. If there are significant deviations between a person's hand actions and the standard assembly actions, establish pose anomalies to identify whether there are incorrect operations by the person.

[0036] Use the hand pose and fingertip motion modeling to identify collaborative assembly anomalies in the pressure data set, combine multiple data sources to analyze whether there are problems in the assembly process, and thus establish assembly anomalies. For example, if it is detected that the hand pose of a person when pressing the door lock component is correct, but the fingertip pressure is lower than 30 N, it means that the person may not apply enough force to fix the component, and it is determined as an assembly anomaly.

[0037] Finally, based on the analysis results of posture anomalies and assembly anomalies, a worker assembly dataset is constructed, which contains all the key data of the worker during the entire assembly process.

[0038] Furthermore, using the hand pose and fingertip motion modeling to perform collaborative assembly anomaly recognition on the pressure dataset, assembly anomalies are established, including: performing time series alignment on the hand pose and fingertip motion modeling and the pressure dataset; extracting pressure features from the pressure dataset, and establishing pressure anomaly nodes according to the pressure feature extraction results; invoking the hand pose and fingertip motion modeling after time series alignment through the pressure anomaly nodes, and performing collaborative recognition of assembly pressing based on pose accuracy according to the invocation results to establish assembly anomalies.

[0039] Specifically, the hand pose and fingertip motion modeling from different data sources and the pressure dataset are synchronously processed in chronological order so that they can be analyzed on the same time axis.

[0040] Extract pressure features from the pressure dataset, such as the maximum value, minimum value, change rate, and stability of the pressure, to obtain the pressure feature extraction results. Establish pressure anomaly nodes according to the pressure feature extraction results. A pressure anomaly node refers to an abnormal pressure state detected at a specific time point or under specific circumstances, that is, a node representing the applied pressure, which is used for joint recognition of fingertip motion and hand posture according to the time node of the applied pressure.

[0041] Invoking the hand pose and fingertip motion modeling after time series alignment through the pressure anomaly nodes, and performing collaborative recognition of assembly pressing based on pose accuracy according to the invocation results to establish assembly anomalies means that the system needs to Use the time node of the pressure anomaly to analyze the motion state of the hand, and combine the hand pose and fingertip motion modeling to identify abnormal situations during the assembly process, that is, combine the posture accuracy of the hand and the pressing force to analyze the assembly quality. For example, if it is detected that the wrist angle deviates from the standard value by more than 5 degrees when a person presses a component and the pressure is unstable, it can be judged that this operation may cause the component to fail to be fixed correctly, and then it is marked as an assembly anomaly.

[0042] Furthermore, as Figure 2As shown in the figure, the collaborative analysis module 40 includes: a batch analysis sub-module 41, which is used to call the batch collaborative analysis layer, perform assembly deviation fitting analysis under the same batch based on the detection data set, and establish a batch anomaly recognition result; a depth analysis sub-module 42, which is used to call the depth collaborative analysis layer, extract the target detection data of the car door lock numbers in the detection data set, and perform joint assembly anomaly analysis according to the target detection data, the multi-dimensional linkage data set, and the worker assembly data set, and establish a depth anomaly recognition result; an output module 43, which is used to receive the batch anomaly recognition result and the depth anomaly recognition result, and construct the anomaly recognition result.

[0043] Specifically, the batch analysis sub-module 41 calling the batch collaborative analysis layer means that this sub-module relies on a computational layer specifically for batch data analysis, which can process a large amount of data simultaneously and find the patterns among them. The detection data set refers to all the detection data collected during the assembly process, such as pressure data, hand trajectory data, assembly time, etc. The focus of batch analysis is on the assembly deviation fitting analysis under the same batch. Assembly deviation refers to the error between the actual assembly situation and the standard assembly, and fitting analysis is to calculate the trend of these errors using mathematical methods. Through batch analysis, these abnormal data can be found and the batch anomaly recognition result can be established, which can effectively screen out the assembly behaviors with large deviations from the standard and improve the overall assembly quality.

[0044] The depth analysis sub-module 42 calling the depth collaborative analysis layer means adopting a higher-precision calculation method to ensure the accuracy of anomaly detection. The target detection data of the car door lock numbers refers to all the detection information related to a specific door lock. The multi-dimensional linkage data set refers to the information combining multiple data sources, such as worker assembly data, detection data, etc., while the worker assembly data set is specifically used to record the operation data of workers during the assembly process, including gestures, pressing force, etc. The purpose of joint assembly anomaly analysis is to integrate these data and find deeper assembly anomalies. For example, if it is found during the detection of a certain car door lock that the locking force is insufficient, and after in-depth analysis, it is found that the hand trajectory of the assembler of this door lock deviates from the standard range during operation, it can be inferred that the reason for the insufficient locking force may be the incorrect hand posture during the assembly process. Through in-depth analysis, a more accurate anomaly recognition result can be established, making the detection more intelligent.

[0045] The role of the output module 43 is to receive the batch anomaly recognition result and the depth anomaly recognition result, and finally construct the anomaly recognition result. The batch anomaly recognition result is the overall anomaly situation obtained based on the assembly data of the same batch, while the depth anomaly recognition result is a detailed analysis of the specific assembly process. The output module 43 combines these two types of data to form the final anomaly recognition report. The final anomaly recognition result can help the production line optimize the assembly process, reduce the defective rate, and improve the overall assembly quality.

[0046] Further, the output module 43 includes: an additional analysis sub-module, which takes the detection node of the car door lock number as the time zero point, performs data interception of the detection data set that is symmetric about the zero point, establishes an additional verification window, performs assembly verification based on the additional verification window, generates an additional anomaly recognition result, and constructs the anomaly recognition result after combining the additional anomaly recognition result, the batch anomaly recognition result, and the in-depth anomaly recognition result.

[0047] Specifically, the function of the output module 43 is to integrate the anomaly recognition results at different levels to ensure the integrity and accuracy of anomaly detection. Among them, the role of the additional analysis sub-module is to further optimize anomaly recognition. Through time-symmetric data interception and assembly verification, the anomaly detection becomes more accurate. By establishing an additional verification window, the detection data is further analyzed, and combined with the results of other analysis modules, a complete anomaly recognition report is finally formed.

[0048] During the detection process of the additional analysis sub-module, each car door lock has a unique number, and the time point corresponding to this number is used as the reference point during detection. For example, if the assembly completion time point of a certain car door lock is set as the zero point, all relevant data will be calculated relative to this time point. Then, data interception of the detection data set that is symmetric about the zero point is performed, and the detection data within a certain time range before and after this zero point is intercepted to analyze data such as the pressure, torque, and position of the door lock to ensure the integrity of the assembly process, avoid single-point data errors, and improve the accuracy of detection.

[0049] After the data interception is completed, an additional verification window is established to provide more accurate assembly verification and generate an additional anomaly recognition result. For example, if the standard assembly time for a certain door lock is from 45 seconds to 55 seconds, and a certain worker takes 60 seconds to complete the assembly, this situation is marked as an anomaly. However, if it is found in the additional verification window that the worker performed additional correction operations within the last 5 seconds, making the assembly quality reach the qualified standard, then false alarms can be avoided.

[0050] Finally, after combining the additional anomaly recognition result, the batch anomaly recognition result, and the in-depth anomaly recognition result, the anomaly recognition result is constructed, indicating that the final anomaly recognition result is jointly determined by multiple data sources. The batch anomaly recognition result is based on the statistical analysis of the entire batch of data and can detect deviations in the overall assembly trend; the in-depth anomaly recognition result is based on the assembly situation of a specific door lock, and through the joint analysis of multi-dimensional data, more detailed anomalies are found; the additional anomaly recognition result is based on the verification of the assembly process within a specific time range to ensure that the detection result will not be misjudged due to data fluctuations at a single time point. By integrating the three types of anomaly recognition results, a more comprehensive anomaly recognition report is generated.

[0051] Furthermore, the multi-dimensional collaborative detection module 20 includes: a focus configuration module, which is used to reconstruct the attention of data indicators according to the anomaly analysis result, establish the result of the reconstructed attention of indicators, use the result of the reconstructed attention of indicators to identify the principal components, reshape the indicators based on the result of the principal component identification, and reallocate the indicator weights based on the result of the reconstructed attention of indicators to complete the linkage acquisition focus configuration.

[0052] Specifically, the multi-dimensional collaborative detection module 20 includes a focus configuration module, which is used to dynamically adjust the data indicators so that it can optimize the data analysis strategy according to different situations to improve the detection accuracy. Among them, the main task of the focus configuration module is to reconstruct the attention of data indicators according to the anomaly analysis result, and then establish the result of the reconstructed attention of indicators. The attention of data indicators represents the degree of emphasis on different detection items. For example, in some cases, the pressure data may be more important than the hand gesture, while in other cases, the hand trajectory may require more attention.

[0053] Using the result of the reconstructed attention of indicators to identify the principal components means that after the attention is adjusted, further analyze the key variables in the data, find out the main factors affecting the detection result, and obtain the result of the principal component identification. For example, when detecting the quality of a car door lock, multiple data dimensions may be involved, such as pressure, time, hand trajectory, etc. The principal component identification can analyze these data to find out the key factors that most affect the assembly quality.

[0054] Reshaping the indicators based on the result of the principal component identification means redefining the detection indicators according to the analysis result of the principal component identification to obtain the result of the reconstructed attention of indicators. Reshaping the indicators means adjusting the configuration of the detection parameters to make it more in line with the actual detection needs. For example, if it is found that the key factor affecting the assembly anomaly is the hand trajectory rather than the traditional pressure, the detection accuracy of the hand trajectory can be increased, and the dependence on the pressure data can be reduced, making the detection more accurate, reducing the possibility of misjudgment, and improving the intelligent level.

[0055] Reallocating the indicator weights based on the result of the reconstructed attention of indicators, readjusting the attention weights of each indicator to complete the linkage acquisition focus configuration, and finally determining a new detection strategy. The indicator weight refers to the importance of different detection parameters in the analysis. Completing the linkage acquisition focus configuration enables it to capture key anomalies more accurately, improving the reliability and intelligence of the detection.

[0056] Furthermore, this application also includes: an anomaly discrimination module, which is used to perform anomaly discrimination of group anomalies and individual anomalies according to the anomaly recognition result, and establish the anomaly discrimination result; an early warning reporting module, which is used to perform anomaly level matching based on the anomaly discrimination result and perform early warning reporting according to the anomaly level matching result.

[0057] Specifically, the anomaly discrimination module is used to discriminate the detected anomalies and issue warnings when necessary to ensure the stability and consistency of the assembly quality. Among them, by distinguishing the scope and impact degree of the anomalies, and the warning reporting module is responsible for taking corresponding countermeasures according to the severity of the anomalies, that is, conducting anomaly discrimination for group anomalies and individual anomalies based on the anomaly recognition results. For example, during the assembly process of a batch of door locks, if multiple workers all have similar assembly deviations at a specific step, such as low pressing force, insufficient torque, etc., it may be judged as a group anomaly. Such anomalies may be caused by factors such as equipment calibration problems, errors in process guidance documents, or insufficient training, so overall adjustments are required.

[0058] On the other hand, if only individual workers have anomalies during the assembly process while the operations of other workers are basically normal, this situation will be determined as an individual anomaly. Individual anomalies are usually related to the operation habits, fatigue level, or individual capabilities of the workers. For example, among 100 workers, only 2 workers have a pressing force lower than 50 N when installing door locks, while the operations of other workers meet the standards, then this will be identified as an individual anomaly, and it may be recommended to provide additional training to these 2 workers or remind them to pay attention to operation details. When the anomaly discrimination module completes the classification of group anomalies and individual anomalies, an anomaly discrimination result is established.

[0059] The warning reporting module performs anomaly level matching based on the anomaly discrimination result. Anomaly level matching means dividing the anomalies into different levels according to the severity of the anomalies, such as low-level, medium-level, and high-level anomalies. For example, if an anomaly is just a minor error that occasionally occurs to a worker and does not affect the final assembly quality, it may be marked as a low-level anomaly; if the anomaly causes a decline in product quality but can be corrected by rework, it may be marked as a medium-level anomaly; if the anomaly seriously affects product safety or results in the disqualification of the entire batch of products, it will be marked as a high-level anomaly. For example, if it is detected that the closing force of the door locks on a certain production line is low and this anomaly involves most of the products, it may be determined as a high-level anomaly and immediate measures need to be taken.

[0060] After the anomaly level matching is completed, warnings are issued based on the matching results. Warning issuance means that the system takes corresponding warning and handling measures according to the severity of the anomaly. For example, if a slight anomaly is detected in a worker's assembly action, the worker may be reminded to adjust the operation through a warning light or screen message; if a quality risk is detected in a batch of products, an alarm may be automatically sent to the quality inspection personnel for additional inspection; if it is found that a certain device or process parameter causes a large-scale anomaly, automatic production suspension may be triggered and engineers notified for an emergency inspection. For example, if a large number of anomalies are detected in the motor installation of a production line, three alarms may be continuously issued within 5 minutes, and the production line will be automatically suspended after the number of anomalies exceeds the set threshold.

[0061] Furthermore, this application also includes: In the general detection module 30, the general detection of the car door lock includes the detection of the dimensional accuracy of components, the detection of the fit clearance, the detection of the micro-motor parameters, the detection of the sensing function, the detection of the quality of the circuit connection, and the detection of the locking and unlocking actions.

[0062] Specifically, the general detection module 30 conducts general detection on the car door lock to ensure that it meets the design standards and usage requirements. Among them, multiple key parameters of the car door lock are detected to identify potential assembly defects or performance anomalies.

[0063] First of all, the general detection module 30 performs the detection of the dimensional accuracy of components. Among them, the components refer to the various structural parts that make up the car door lock, such as the lock case, lock hook, spring, transmission mechanism, etc. The dimensional accuracy detection means measuring whether the specific dimensions of these components are within the specified tolerance range. If the dimensions exceed the range, it may cause the door lock to not close properly or become loose, thus affecting the usage safety.

[0064] Secondly, the general detection module 30 also needs to conduct the detection of the fit clearance. The fit clearance refers to the reserved gap between two components, which affects the smooth movement and sealing performance of the door lock. For example, if the gap between the lock tongue and the lock catch is too large, it may cause abnormal noises in the door lock during driving; if the gap is too small, it may cause the door lock to be difficult to close. Therefore, the fit clearance of key parts is measured and ensured to be within a reasonable range.

[0065] Moreover, the general detection module 30 needs to perform the detection of the micro-motor parameters. The micro-motor is responsible for driving the lock tongue to complete the unlocking or locking action. The micro-motor parameters include working current, rotational speed, torque, etc. For example, if the motor current of a certain door lock increases abnormally, it may mean that the motor has an overload problem, which may be caused by excessive assembly friction or an internal motor fault. Therefore, detecting these parameters helps to determine whether the door lock can work properly and ensure its durability.

[0066] Next, the sensing function needs to be detected. The car door lock is equipped with various sensors, such as position sensors, Hall sensors, etc., which are used to detect the opening and closing state of the door lock and send signals to the vehicle control system. The sensing function detection refers to checking whether the sensors can accurately sense the state of the door lock. For example, a certain intelligent door lock needs to send a signal to the central control system immediately after unlocking to trigger the door unlocking function. If the sensor fails, it may cause the vehicle to fail to detect the unlocked state, thus affecting the normal opening of the door. Therefore, different operation scenarios are simulated and the response of the sensors is verified.

[0067] In addition, the general detection module 30 is also involved in the detection of the quality of line connections. The circuit system of the car door lock is responsible for connecting the micro-motor, sensors and control unit, and the detection of the quality of line connections is mainly used to check whether the electrical connections are reliable. For example, if the circuit welding is not firm, it may cause the micro-motor to break in a vibrating environment, thus resulting in the failure of the door lock. Therefore, by means of resistance measurement, conductivity test, etc., ensure the stability of the line connection.

[0068] Finally, the general detection module 30 needs to conduct the locking and unlocking action detection. The locking and unlocking action means that after receiving a control signal or a mechanical operation, the door lock can smoothly complete the locking or unlocking process. For example, in the automatic locking mode, the door lock should automatically lock when the vehicle reaches a specific speed, and when the owner presses the unlock button, the door lock should immediately unlock. The locking and unlocking action detection is to check the response speed, running smoothness and whether there is jamming of the door lock through automated testing or manual operation.

[0069] Furthermore, the present application further includes: a recording module, which is used to record the abnormal recognition results, establish a sensitive window according to the abnormal recognition results, and use the sensitive window to conduct sensitive monitoring management after abnormal warning.

[0070] Specifically, the function of the recording module is to store the abnormal recognition results, further optimize the monitoring ability of the system, and improve the accuracy of abnormal detection.

[0071] First of all, the recording module needs to record the abnormal recognition results. Among them, the detected abnormal situations are stored in a database or a log system to ensure that they can be analyzed and traced later. The abnormal recognition results refer to assembly deviations, functional abnormalities or other quality problems found during the detection process. By attaching specific abnormal categories and detection data for subsequent analysis.

[0072] Then, a sensitive window is established based on the anomaly recognition result. The sensitive window is a time or data range setting used to focus on the time period or product batch where anomalies may occur. The sensitive window can be time-based, for example, an abnormal increase is detected during a certain period of a certain day; or it can be data feature-based, for example, the assembly actions of a certain operator deviate significantly from the standard actions, resulting in an increase in the anomaly rate.

[0073] Next, sensitive monitoring management is carried out using the sensitive window after anomaly warning. Anomaly warning means that after detecting an abnormal situation, an alarm is sent to relevant personnel or equipment in a timely manner so that corrective measures can be taken. For example, if the pressure data is abnormal during the assembly process of a certain door lock product, an alarm is immediately sent to the production line management personnel to remind them to check the assembly process or equipment status. Sensitive monitoring management refers to further monitoring and analysis of the production situation within the sensitive window after warning. For example, after identifying an anomaly, the detection strategy is adjusted, such as increasing the sampling ratio of this batch of door locks or adding additional detection items to ensure that the problem does not continue to expand.

[0074] Furthermore, this application also includes an adaptive correction module, which is used to receive the detection feedback from the user, verify the anomaly recognition result according to the detection feedback, establish a correction feedback, and optimize the collaborative analysis module 40 according to the correction feedback.

[0075] Specifically, the role of the adaptive correction module is to verify the anomaly recognition result based on the user feedback and optimize the collaborative analysis module 40 accordingly to improve the detection accuracy and adaptability.

[0076] First of all, the adaptive correction module needs to receive the detection feedback from the user. Among them, the user can provide data through manual input, sensor device feedback or automatic recording. The user refers to inspectors, assembly workers or production management personnel who may find errors in the anomaly recognition result when using the system. For example, if the system detects that there is a locking anomaly in a certain car door lock, but the inspector finds that the door lock function is normal after actual inspection, the inspector can submit feedback indicating that this recognition result may be a false alarm. Detection feedback refers to the feedback information provided by the user regarding the recognition accuracy, including whether the anomaly is real, whether the anomaly category is correct, and whether the detection standard is reasonable.

[0077] Next, the anomaly recognition result is verified according to the detection feedback. Recognition verification refers to rechecking the anomaly recognition result to confirm whether the judgment is accurate. For example, if the assembly anomaly rate of a certain batch of door locks is higher than normal, and this batch is marked as abnormal, but the user feedback indicates that some detection results may be in error, then these results need to be reanalyzed to check whether false alarms are caused by sensor errors, detection standard deviations or algorithm problems.

[0078] Then, establish calibration feedback. Calibration feedback refers to making adjustments to the parts with recognition errors based on the information provided by the user and the results of the system's self-check. For example, if a high false alarm rate is found, the abnormal discrimination threshold can be adjusted to reduce the false judgment of normal door locks; if it is found that some abnormal types cannot be recognized, new detection rules can be added to improve the recognition ability. For example, if the inspector feedbacks that the system is prone to misjudge a certain type of door lock during the fit clearance detection, the model can be retrained using historical data to make it more suitable for the characteristics of this type of door lock.

[0079] Finally, optimize the collaborative analysis module 40 according to the above calibration feedback. The optimization of the collaborative analysis module 40 means using the corrected data to further adjust the abnormal recognition algorithm to improve the overall intelligence level of the system. For example, if in 10,000 detections, the initial false alarm rate of the system is 5%, and after optimization through calibration feedback, the false alarm rate is reduced to 2%, it indicates that the abnormal recognition ability of the collaborative analysis module 40 has been improved. The optimization methods may include adjusting data weights, optimizing deep learning models, resetting the parameter range of abnormal detection, etc. For example, if a certain error pattern is found in the angular velocity data of the inertial measurement unit, the model can be trained using new data samples to more accurately identify abnormal assembly operations.

[0080] In summary, the dynamic quality monitoring system for intelligent assembly detection of automotive door locks provided by this application has the following technical effects: By achieving the technical goals of multi-dimensional real-time monitoring, intelligent analysis, and accurate abnormal recognition of the automotive door lock assembly process, it achieves the technical effects of improving assembly quality, increasing production efficiency, reducing rework costs, and enhancing the intelligence level of the manufacturing process.

[0081] The above description of the disclosed embodiments enables those skilled in the art to implement or use this application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application will not be limited to the embodiments shown herein, but rather will be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0082] Obviously, those skilled in the art can make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of this application and its equivalent technologies, this application is also intended to include these changes and modifications.

Claims

1. A dynamic quality monitoring system for intelligent assembly detection of automotive door locks, characterized in that, The system includes: An intelligent tactile glove for recording a worker assembly data set with an automotive door lock number identifier after signal interaction with an automotive door lock, and the intelligent tactile glove is integrated with a flexible pressure sensor and an inertial measurement unit; A multi-dimensional collaborative detection module for performing abnormality analysis after receiving the worker assembly data set, configuring linkage collection attention according to the abnormality analysis result, and performing assembly linkage collection based on the linkage collection attention after detecting the corresponding automotive door lock number to establish a multi-dimensional linkage data set; A general detection module for performing general detection on the automotive door lock to establish a detection data set; A collaborative analysis module for calling a collaborative analysis network to perform assembly quality abnormality identification based on the detection data set, the multi-dimensional linkage data set, and the worker assembly data set to establish an abnormality identification result, and the collaborative analysis network includes a batch collaborative analysis layer and a depth collaborative analysis layer.

2. The dynamic quality monitoring system for intelligent assembly detection of automotive door locks according to claim 1, wherein, The intelligent tactile glove is used for: Performing monitoring startup after signal interaction with the automotive door lock; Activating the flexible pressure sensors at the fingertips and palms of the intelligent tactile glove to establish a pressure data set; Activating the inertial measurement unit to establish an inertial data set, reconstructing the spatial trajectory based on the action feature extraction channel, resolving the hand pose, and performing fingertip motion modeling; Performing standard action pose matching according to the hand pose and fingertip motion modeling to establish pose abnormality; Performing collaborative assembly abnormality identification on the pressure data set using the hand pose and fingertip motion modeling to establish assembly abnormality; Constructing a worker assembly data set according to the pose abnormality and the assembly abnormality.

3. The dynamic quality monitoring system for intelligent assembly detection of automotive door locks according to claim 2, characterized in that In the intelligent tactile glove, performing collaborative assembly abnormality identification on the pressure data set using the hand pose and fingertip motion modeling to establish assembly abnormality, including: Performing time series alignment on the hand pose and fingertip motion modeling and the pressure data set; Performing pressure feature extraction on the pressure data set and establishing a pressure abnormality node according to the pressure feature extraction result; Calling the hand pose and fingertip motion modeling after time series alignment through the pressure abnormality node, and performing assembly pressing collaboration identification based on pose accuracy according to the call result to establish assembly abnormality.

4. The dynamic quality monitoring system for intelligent assembly detection of automotive door locks according to claim 1, characterized in that, The collaborative analysis module includes: A batch analysis sub-module for calling the batch collaborative analysis layer to perform assembly deviation fitting analysis under the same batch based on the detection data set to establish a batch abnormality identification result; A depth analysis sub-module for calling the depth collaborative analysis layer to extract the target detection data of the automotive door lock number in the detection data set, and performing joint assembly abnormality analysis according to the target detection data, the multi-dimensional linkage data set, and the worker assembly data set to establish a depth abnormality identification result; An output module for constructing the abnormality identification result after receiving the batch abnormality identification result and the depth abnormality identification result.

5. The dynamic quality monitoring system for intelligent assembly detection of automotive door locks according to claim 4, wherein, The output module includes: An additional analysis sub-module, which takes the detection node of the car door lock number as the time zero point, performs data interception of the detection data set that is symmetric about the zero point, establishes an additional verification window, performs assembly verification based on the additional verification window, generates an additional anomaly recognition result, and constructs the anomaly recognition result after combining the additional anomaly recognition result, the batch anomaly recognition result, and the in-depth anomaly recognition result.

6. The dynamic quality monitoring system for intelligent assembly detection of automotive door locks according to claim 1, wherein The multi-dimensional collaborative detection module includes: A focus configuration module, which is used to reconstruct the attention of data indicators according to the anomaly degree analysis result, establish a result of index attention reconstruction, use the result of index attention reconstruction to perform principal component identification, reshape the indicators with the principal component identification result, and re-allocate the index weights based on the result of index attention reconstruction to complete the linkage acquisition focus configuration.

7. The dynamic quality monitoring system for intelligent assembly detection of automotive door locks according to claim 1, characterized in that The system includes: An anomaly discrimination module, which is used to perform anomaly discrimination of group anomalies and individual anomalies according to the anomaly recognition result, and establish an anomaly discrimination result; An early warning reporting module, which is used to perform anomaly level matching based on the anomaly discrimination result and perform early warning reporting according to the anomaly level matching result.

8. The dynamic quality monitoring system for intelligent assembly detection of automotive door locks according to claim 1, characterized in that, In the general detection module, the general detection of the car door lock includes the detection of the dimensional accuracy of parts, the detection of the fit clearance, the detection of the micro-motor parameters, the detection of the sensing function, the detection of the quality of the line connection, and the detection of the locking and unlocking actions.

9. The dynamic quality monitoring system for intelligent assembly detection of automotive door locks according to claim 1, characterized in that, The system includes: A recording module, which is used to record the anomaly recognition result, establish a sensitive window according to the anomaly recognition result, and perform sensitive monitoring management after the anomaly early warning by using the sensitive window.

10. The dynamic quality monitoring system for intelligent assembly detection of automobile door locks according to claim 1, characterized in that The system includes: An adaptive correction module, which is used to receive the detection feedback from the user, perform recognition verification of the anomaly recognition result according to the detection feedback, establish a correction feedback, and optimize the collaborative analysis module according to the correction feedback.

Images