One-stop offline detection method for automobile tail door stay bar

Abstract

The invention discloses a one-stop offline detection method for an automobile tail door stay bar, and relates to the technical field of automobile tail door stay bar detection, and the method specifically comprises the following steps: carrying out the real-time detection of the process of the telescoping push-pull circulation operation of the automobile tail door stay bar, so as to judge whether the push-pull motion is coherent and damaged; when the phenomenon that the push-and-pull action is coherent and damaged is detected, the whole-process characteristic information of the push-and-pull action of the stay bar is collected in real time through a deployed sensor network, and the whole-process characteristic information of the push-and-pull action of the stay bar is preprocessed; analyzing the preprocessed whole-process characteristic information, and evaluating the damage degree when a push-pull action coherent damage phenomenon occurs; and respectively executing corresponding measures according to the assessment result of the damage degree. According to the invention, the problem that layered screening and response cannot be carried out according to the damage degree of the coherent damage phenomenon of the push-pull action in the existing one-stop offline detection of the automobile tail door stay bar is solved, and real-time detection, quantitative evaluation and intelligent grading processing of the abnormal stay bar are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of automotive tailgate strut detection, and particularly to a one-stop offline detection method for automotive tailgate struts. Background Art

[0002] The one-stop offline detection of automotive tailgate struts refers to, after the production of automotive tailgate struts, through an integrated and automated detection system, continuously completing a comprehensive detection of various performance indicators of the struts (such as push-pull force, stroke length, airtightness, durability, etc.) on a single detection line, and immediately generating detection data and a pass / fail determination to ensure that the product meets the design standards and usage requirements before leaving the factory. Since the automotive tailgate strut is an important component for supporting the opening and closing of the tailgate, its performance is directly related to vehicle safety and user experience. Therefore, it is necessary to conduct strict and comprehensive detection on it during the offline process. Traditional step-by-step detection has problems such as scattered detection links, low detection efficiency, excessive manual intervention, and easy errors in data recording, making it difficult to meet the requirements of modern automotive manufacturing for efficient, accurate, and traceable quality management. Through one-stop offline detection, the detection efficiency can be greatly improved, the errors caused by manual operations can be reduced, the consistency and integrity of detection data can be ensured, and at the same time, the production line rhythm can be accelerated and the overall production cost can be reduced. Thus, while improving the quality stability of the strut product, the market competitiveness of the enterprise can be enhanced. Therefore, researching and applying the one-stop offline detection method for automotive tailgate struts is an important means to improve the manufacturing process level of struts, realize intelligent manufacturing, and achieve quality controllability.

[0003] The existing one-stop offline detection technology for automotive tailgate struts usually includes clamping and positioning, push-pull force detection, stroke detection, airtightness detection, durability detection, appearance detection, data recording and judgment, etc. The entire process is highly integrated on an automated detection line. First, a dedicated fixture quickly and accurately positions and fixes the strut to be tested to ensure the standardization of the detection process. Subsequently, the push-pull force detection device performs telescopic actions on the strut and measures the push and pull force values in real time to determine whether they meet the design parameters. Immediately afterwards, the stroke detection is carried out. The sensor monitors the stroke length when the strut is fully extended and retracted to ensure that the mechanical characteristics of the strut are within the specified range. The airtightness detection applies a certain air pressure inside the strut and monitors the leakage rate to determine whether the sealing performance of the strut meets the standard. The durability detection usually adopts a rapid cyclic telescopic method to conduct a certain number of life tests on the strut to verify its reliability during long-term use. At the same time, the system conducts an appearance inspection on the surface of the strut to identify defects such as scratches and dents. All detection data is automatically collected and recorded through an industrial control system, and automatically judged as qualified or unqualified according to the preset standards. The unqualified products are sorted and processed to achieve the standardized, automated and traceable management of the whole process. Through such a detection process, the existing one-stop offline detection technology can ensure the stable and reliable performance and quality of each strut leaving the factory while guaranteeing a high detection speed.

[0004] The prior art has the following deficiencies: In the durability push-pull action detection link of the one-stop offline detection process for tailgate struts, the detection system conducts continuous multiple telescopic push-pull cycles on the strut to evaluate the stability of its service life. In the actual detection process, due to the loss of lubricant inside the strut after multiple cycles, and at the same time, the friction force distribution becomes uneven due to increased wear on the component contact surfaces, resulting in a sudden slowdown or jamming of the push-pull speed during the push-pull action, and the continuity of the push-pull action is damaged to varying degrees. However, the existing one-stop offline detection technology for automotive tailgate struts cannot execute corresponding measures according to the degree of damage when the continuity of the push-pull action is damaged in the durability push-pull action detection link, resulting in the inability to timely identify and hierarchically screen when there is an abnormal trend in the action continuity, allowing abnormal struts with different degrees of damage to continue to flow into the subsequent assembly process as qualified products, thus leading to quality risks such as unsmooth tailgate opening and closing, shortened strut life, increased batch after-sales complaints, and large-scale warranty claims in actual use.

[0005] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure, and thus it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0006] The object of the present invention is to provide a one-stop offline detection method for automotive tailgate struts to solve the problems in the above-mentioned background technology.

[0007] To achieve the above object, the present invention provides the following technical solution: A one-stop offline detection method for automotive tailgate struts, specifically including the following steps: Real-time detection is carried out on the process of the telescopic push-pull cyclic operation of the automotive tailgate strut to determine whether there is a phenomenon of impaired coherence of the push-pull action. When the phenomenon of impaired coherence of the push-pull action is detected, the whole-process characteristic information of the strut push-pull action is collected in real time through the deployed sensor network and preprocessed. Analyze the preprocessed whole-process characteristic information to evaluate the degree of impairment when the phenomenon of impaired coherence of the push-pull action occurs. According to the evaluation result of the degree of impairment, corresponding measures are respectively executed. Store the detection data, evaluation results and processing records in the database, and continuously optimize the detection strategy and quality trend analysis based on historical data.

[0008] Preferably, analyzing the preprocessed whole-process characteristic information to evaluate the degree of impairment when the phenomenon of impaired coherence of the push-pull action occurs specifically includes the following steps: Extract the motion behavior disturbance information and dynamic execution deviation information from the preprocessed whole-process characteristic information, and analyze them after extraction to generate an action fluctuation coefficient and a mode deviation index respectively. Construct an impairment degree evaluation model for the generated action fluctuation coefficient and mode deviation index, and generate an impairment coefficient through weighted summation. Determine the preset impairment coefficient threshold interval, and compare it with the generated impairment coefficient after determination, and evaluate the degree of impairment when the phenomenon of impaired coherence of the push-pull action occurs according to the comparison result.

[0009] Preferably, the acquisition logic of the action fluctuation coefficient is as follows: Extract the motion behavior disturbance information from the preprocessed whole-process characteristic information, specifically including the actual displacement of the automotive tailgate strut at different times within a period when the phenomenon of impaired coherence of the push-pull action occurs and the actual push-pull speed acting on the automotive tailgate strut, and respectively calibrate them as and , represents the actual displacement of the automotive tailgate strut at time within a period when the phenomenon of impaired coherence of the push-pull action occurs, represents the actual push-pull speed acting on the automotive tailgate strut at time within a period when the phenomenon of impaired coherence of the push-pull action occurs, , is a positive integer; Calculate the average value of the actual displacement of the vehicle tailgate strut at all times within a period when the phenomenon of impaired continuity of pushing and pulling actions occurs , according to the formula: ; Calculate the average value of the actual pushing and pulling speed acting on the vehicle tailgate strut at all times within a period when the phenomenon of impaired continuity of pushing and pulling actions occurs , according to the formula: ; Calculate the action fluctuation coefficient, and the specific calculation formula is as follows: In the formula, is the action fluctuation coefficient.

[0010] Preferably, the acquisition logic of the mode deviation index is as follows: Extract the dynamic execution deviation information from the preprocessed full-process characteristic information, specifically including the actual acceleration of the vehicle tailgate strut and the pushing and pulling resistance received by the vehicle tailgate strut at different times within a period when the phenomenon of impaired continuity of pushing and pulling actions occurs, and respectively calibrate them as and , represents the actual acceleration of the vehicle tailgate strut at the th moment within a period when the phenomenon of impaired continuity of pushing and pulling actions occurs, represents the pushing and pulling resistance received by the vehicle tailgate strut at the th moment within a period when the phenomenon of impaired continuity of pushing and pulling actions occurs, , is a positive integer; Through cluster analysis and statistical modeling of the acceleration and pushing and pulling resistance data of historical qualified strut samples at each moment, determine the standard acceleration reference value and the standard pushing and pulling resistance reference value, and respectively calibrate them as and ; Calculate the mode deviation index, and the specific calculation formula is as follows: In the formula, is the mode deviation index.

[0011] Preferably, for the generated action fluctuation coefficient and the mode deviation index construct a damage degree evaluation model, and generate a damage coefficient through weighted summation. The specific calculation formula is as follows: In the formula, is the damage coefficient, and are the non-zero weight coefficients of the action fluctuation coefficient and the mode deviation index respectively, and 。

[0012] Preferably, a preset damaged coefficient threshold range is determined , and after determination, it is compared with the generated damaged coefficient to evaluate the degree of damage when the phenomenon of impaired push-pull action coherence occurs according to the comparison result. The specific comparison and analysis are as follows: If , the degree of damage when the phenomenon of impaired push-pull action coherence occurs is low; If , the degree of damage when the phenomenon of impaired push-pull action coherence occurs is medium; If , the degree of damage when the phenomenon of impaired push-pull action coherence occurs is severe.

[0013] Preferably, according to the evaluation result of the degree of damage, corresponding measures are respectively executed. Specifically: If the evaluation result shows that the degree of damage is low, the specific measure to be executed is: directly determine the corresponding support rod as qualified and allow it to enter the subsequent assembly process; If the evaluation result shows that the degree of damage is medium, the specific measure to be executed is: perform a re-inspection process on the corresponding support rod, conduct repeated detection and supplementary measurement of action characteristics, and re-classify and process according to the re-inspection result; If the evaluation result shows that the degree of damage is severe, the specific measure to be executed is: determine the corresponding support rod as a non-conforming product and remove it from the assembly process, and enter the non-conforming product processing station for isolation.

[0014] In the above technical solution, the technical effects and advantages provided by the present invention are: 1. In the one-stop offline detection process of the automotive tailgate support rod, the present invention can detect the whole-process characteristic information of the push-pull action in real time, and automatically trigger the data acquisition and preprocessing process when the phenomenon of impaired push-pull action coherence is detected. The present invention can capture the abnormal push-pull action caused by factors such as lubricant loss and component wear in the first time, avoiding the problem that traditional detection technologies only rely on the compliance of the number of cycles and ignore the action details. Through fine real-time detection, the present invention greatly improves the early discovery ability of the potential coherence deterioration trend of the support rod, enhances the accuracy and timeliness of abnormal recognition, and lays a reliable foundation for subsequent hierarchical screening and quality control.

[0015] 2. By extracting motion behavior perturbation information and dynamic execution deviation information, calculating the action fluctuation coefficient and the pattern deviation index respectively, and generating a damage coefficient based on weighted summation, the present invention realizes the quantitative evaluation of the impairment degree of the pushing and pulling action coherence. Then, by comparing with the preset damage coefficient threshold interval, the impairment degree is subdivided into low degree, medium degree and severe degree, corresponding to different processing strategies respectively. Compared with the prior art that cannot distinguish different impairment levels and abnormal struts are likely to be mixed into qualified products, the present invention can implement automatic release, re-inspection or rejection measures according to different degrees, realizing the intelligent grading management of the strut quality, significantly reducing the missed inspection rate and misjudgment rate, and enhancing the quality control accuracy of the detection link.

[0016] 3. The present invention also stores the detection data, evaluation results and processing records in the database uniformly, and combines historical data for quality trend analysis and detection strategy optimization, which not only realizes the traceability of abnormal cases and problem tracing, but also supports the dynamic adjustment and continuous evolution of the detection model. By periodically updating the weight parameters of the action fluctuation coefficient and the pattern deviation index and optimizing the damage coefficient threshold setting, the system can continuously adapt to actual working conditions changes such as the changes in the manufacturing batches of struts and the differences in material aging characteristics, significantly improving the long-term stability, accuracy and self-learning ability of the detection system, and effectively supporting the continuous improvement of the product quality system of the tailgate strut and the improvement of customer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings.

[0018] Figure 1 It is a schematic flow chart of a one-stop offline detection method for an automotive tailgate strut according to the present invention.

[0019] Figure 2 It is a method mind map of a one-stop offline detection method for an automotive tailgate strut according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] Now, the exemplary embodiments will be described more fully with reference to the accompanying drawings. However, the exemplary embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these exemplary embodiments are provided so that the present disclosure will be more complete and comprehensive, and will fully convey the concept of the exemplary embodiments to those skilled in the art.

[0021] The present invention provides as Figure 1 and Figure 2A one-stop offline detection method for a car tailgate strut is as follows: During the process of performing telescopic push-pull cyclic operations on the car tailgate strut, real-time detection is carried out to determine whether there is a phenomenon of impaired coherence in the push-pull actions. To achieve real-time detection of the car tailgate strut during the telescopic push-pull cyclic operation process, the system can integrate multiple types of sensors on the detection platform, including displacement sensors, acceleration sensors, action time acquisition modules, push-pull resistance sensors, etc., to construct a sensor network and collect multi-dimensional physical characteristic data of the strut during each cyclic action process. The output signals of all sensors are uploaded to the data processing module in real time through the communication interface. The data processing module runs in the upper computer system, performs time synchronization, caching, and formatting processing on the received data at a high frequency, and stores the data continuously in a time series manner. This multi-dimensional, continuous, and high-frequency data acquisition process can completely reproduce the action process of each cycle, thereby realizing real-time and full-process digital detection of the entire push-pull operation behavior.

[0022] After obtaining the full-process detection data of the strut, the software system analyzes the data of each action cycle by constructing a feature extraction algorithm, and focuses on identifying the following types of performances: whether there is a sudden drop in the action speed during continuous push-pull, whether the action curve is interrupted, whether there are obvious discontinuity points in the acceleration change curve, whether there are abnormal fluctuations in the push-pull resistance within the cycle, etc. The system makes a logical judgment on the abnormal behaviors in the above multiple dimensions and sets a condition threshold for triggering "impaired coherence in the push-pull action", for example: the action speed suddenly drops by more than the set deviation within a certain cycle, there is a pause section in the displacement-time curve, the number of consecutive abnormal actions reaches the set number, etc. Once any one or more of these conditions are met, it can be automatically determined that the strut has shown a phenomenon of impaired coherence in the push-pull action, and then enters the downstream detailed analysis process.

[0023] The reason for carrying out real-time detection and automatic judgment on the action coherence of the strut during the telescopic push-pull operation process is that traditional durability detection methods mainly focus on whether the number of push-pull cycles meets the standard, while ignoring the changes in action quality during the cycle. In particular, coherence anomalies caused by factors such as lubricant consumption and component wear often gradually appear in the middle and late stages of the detection. Although these anomalies still meet the standard in terms of quantity indicators, they have actually shown an early failure trend and are extremely likely to cause faults such as tailgate jamming, jerks, or poor suction during subsequent use. Therefore, by using software to monitor the entire action process in real time and identify the phenomenon of impaired coherence, potential defective products can be discovered earlier, the detection granularity can be improved, and the grading and screening ability can be enhanced, thereby constructing a more rigorous and intelligent quality control system and effectively reducing after-sales risks and quality costs.

[0024] When the phenomenon of impaired continuity of the pushing and pulling actions is detected, the whole-process characteristic information of the pole pushing and pulling actions is collected in real time through the deployed sensor network and preprocessed. In order to collect the whole-process characteristic information of the pole pushing and pulling actions in real time after the phenomenon of impaired continuity of the pushing and pulling actions is detected, the system deploys a sensor network composed of multiple different types of sensors on the detection platform, including but not limited to displacement sensors, acceleration sensors, push-pull force sensors, and time acquisition modules. These sensors are respectively responsible for collecting key data such as displacement changes, motion trends, resistance fluctuations, and action execution cycles during the action process, and establishing a communication connection with the host computer or the edge processing system through wired or wireless data channels. The system realizes the synchronous acquisition of the dynamic parameters at each moment during the whole process of the pushing and pulling actions through software control of the acquisition frequency and trigger mechanism. All the collected original data will be cached in real time in a structured data format and organized and classified according to the action cycle number to ensure that the physical state changes of the whole action process can be accurately restored during the subsequent processing.

[0025] Preprocessing the collected whole-process characteristic information is a key step to ensure the accuracy of subsequent analysis and calculation efficiency. Because the original sensor data often has problems such as signal noise, time axis asynchronization, and inconsistent numerical scales, directly using it will lead to evaluation deviations and even judgment errors. Therefore, the system performs various preprocessing operations on the original data at the software level, including denoising filtering, outlier removal, data normalization, time synchronization correction, discrete point interpolation and reconstruction, etc. Among them, denoising filtering can smooth the high-frequency disturbances through a sliding window or wavelet algorithm; outlier removal judges the mutation points based on the threshold and trend differences and strips them; data normalization unifies the dimension through range normalization or Z-score normalization methods; time synchronization interpolates and aligns the data of each channel according to the master control signal; interpolation reconstruction is used to restore the complete trajectory when there are data gaps. This series of software processing processes can effectively improve the temporal consistency, structural integrity, and physical comparability of the data, ensuring that the subsequent evaluation logic is based on a stable and reliable input.

[0026] Analyze the preprocessed whole-process characteristic information to evaluate the degree of impairment when the phenomenon of impaired continuity of the pushing and pulling actions occurs. In this embodiment, analyzing the preprocessed whole-process characteristic information to evaluate the degree of impairment when the phenomenon of impaired continuity of the pushing and pulling actions occurs specifically includes the following steps: Extract the motion behavior disturbance information and dynamic execution deviation information from the preprocessed whole-process characteristic information, and analyze them after extraction to generate an action fluctuation coefficient and a mode deviation index respectively. At the software implementation level, to extract motion behavior perturbation information and dynamic execution deviation information from the preprocessed full-process characteristic information, it is first necessary to classify the dimensions and map the structures of the preprocessed time series data. Multiclass characteristic data such as displacement, velocity, acceleration, and push-pull resistance corresponding to each moment are respectively reconstructed into a multi-dimensional data matrix under a unified time axis. Subsequently, based on the data label rules and signal feature extraction model, the system automatically filters out the column vectors related to the "displacement-velocity" coupling characteristics for constructing motion behavior perturbation information; at the same time, it extracts the column vectors related to the "acceleration-resistance" coupling mode and its deviation from the historical template for constructing dynamic execution deviation information. During this process, the system will call the built-in data channel screening rules and template alignment mechanism, and through methods such as sliding window analysis, sampling frequency alignment, and curve normalization, realize the registration and extraction of multi-dimensional behavior characteristics and standard data, and finally output a structured perturbation information set and deviation information set, providing feature input for the subsequent parameter calculation module.

[0027] Construct a damage degree evaluation model for the generated action fluctuation coefficient and mode deviation index, and generate a damage coefficient through weighted summation; Determine the pre-set damage coefficient threshold interval, and after determination, compare it with the generated damage coefficient, and evaluate the damage degree when the push-pull action coherence damage phenomenon occurs according to the comparison result.

[0028] To determine the pre-set damage coefficient threshold interval, the system first needs to construct a multi-level damage level label library based on historical sample data. This library consists of a large number of strut detection samples labeled by manual or empirical rules. The samples include the corresponding action fluctuation coefficient, mode deviation index, and actual verification results. Subsequently, the system performs clustering analysis and statistical distribution modeling on this sample set, such as using the Gaussian mixture model (GMM) or K-means clustering algorithm, to fit and segmentally identify the distribution trend of the damage coefficient among different damage levels, and combined with the actual product quality feedback, soft-constraint optimize the interval boundaries. Finally, according to the clustering results and the matching degree of the quality labels, the system automatically generates a multi-segment damage coefficient threshold interval, usually including three levels: mild, moderate, and severe, and writes it into the rule module in the form of configuration parameters for real-time comparison and damage judgment calls, realizing the adaptive setting of the threshold interval and long-term maintainable update.

[0029] In this embodiment, the acquisition logic of the action fluctuation coefficient is as follows: Extract motion behavior perturbation information from the preprocessed full-process characteristic information, specifically including the actual displacement of the vehicle tailgate strut at different moments within a period of time when the push-pull action coherence damage phenomenon occurs and the actual push-pull speed acting on the vehicle tailgate strut, and respectively calibrate them as and , Denote the actual displacement of the vehicle tailgate strut within a period of time when the phenomenon of impaired continuity of pushing and pulling actions occurs at a certain moment, Denote the actual pushing and pulling speed acting on the vehicle tailgate strut within a period of time when the phenomenon of impaired continuity of pushing and pulling actions occurs at a certain moment, , where \(n\) is a positive integer; During the detection of pushing and pulling actions, in order to realize the real-time acquisition of the actual displacement and actual pushing and pulling speed of the vehicle tailgate strut at each moment when the phenomenon of impaired continuity occurs, the detection system constructs a high-precision data acquisition network through a displacement sensor and a speed encoder integrated on the execution device. The displacement sensor can be in the form of an optical encoder, a linear potentiometer or a laser displacement meter, etc., and is used to record the real-time stroke change of the strut during pushing and pulling. Its output signal is synchronously sent to the data processing system through the sampling module, and the software can automatically collect the displacement value corresponding to each moment at a set time interval , forming a continuous displacement time series. At the same time, the pushing and pulling speed is not obtained from a fixed set value, but is automatically deduced by the system according to the change amplitude of the strut displacement within a unit time, or the real-time pushing and pulling speed value is directly output by the speed encoder . After collecting the data, the software system will perform preprocessing such as time series alignment, abnormal rejection and unit conversion, so that the displacement and speed data at each moment can be accurately associated, thus completely recording the real dynamic process of the pushing and pulling actions within a period of time. Among them, the actual displacement denotes the distance change of the strut pushing and pulling stroke relative to the initial position at the th moment, reflecting the completion degree of the mechanical execution; while the actual pushing and pulling speed denotes the instantaneous movement speed of the system applying an action to the strut at this moment, reflecting the rhythm and stability of the pushing and pulling actions. These two types of data, as the core physical basis for action disturbance evaluation, can accurately capture the discontinuous, sudden or fluctuating phenomena occurring during the pushing and pulling process, providing high-quality input for calculating the action fluctuation coefficient

[0030] Calculate the average value of the actual displacement of the vehicle tailgate strut at all moments within a period of time when the phenomenon of impaired continuity of pushing and pulling actions occurs , according to the formula: ; Calculate the average value of the actual pushing and pulling speed acting on the vehicle tailgate strut at all moments within a period of time when the phenomenon of impaired continuity of pushing and pulling actions occurs , according to the formula: ; Calculate the action fluctuation coefficient, and the specific calculation formula is as follows: In the formula, is the action fluctuation coefficient

[0031] In order to accurately reflect the degree of action disturbance during the pushing and pulling process of the automotive tailgate strut, the calculation formula of the action fluctuation coefficient ( ) adopts an exponentially weighted difference cumulative method based on time series. In this formula, first, the actual displacement and the actual pushing and pulling speed at different moments within a period of time when the phenomenon of impaired continuity of the pushing and pulling action occurs are combined and calculated moment by moment. The influence of the displacement value and the speed is coupled through the form of , where the exponential operation is used to strengthen the disturbance effect brought by small displacement fluctuations at high speeds, making the disturbance more sensitive when the pushing and pulling speed changes suddenly; then the coupled value is subtracted from the coupled value of the average displacement and the average speed within the time period to measure the relative deviation between the action state at each moment and the overall stable state, and then the average of the absolute values of the deviations at all moments is calculated to comprehensively evaluate the fluctuation amplitude of the action process; finally, the overall difference is wrapped in the natural logarithm function, which can not only suppress the interference under the low-fluctuation background but also significantly amplify the output when the fluctuation is severe, ensuring that the coefficient has both sufficient sensitivity and numerical stability. This calculation design can effectively capture the disturbance changes caused by jerks, sudden stops, or non-uniform speeds during the pushing and pulling action, thereby providing a quantitative basis for subsequent judgment of impaired continuity.

[0032] The numerical value of the action fluctuation coefficient is positively correlated with the degree of impairment of the continuity of the pushing and pulling action. The core lies in that this coefficient quantifies the dynamic disturbance degree of the strut during the pushing and pulling process. When the automotive tailgate strut runs smoothly during the telescopic process and the speed and displacement changes are stable, the difference between each moment and the overall average state is small, and the cumulative disturbance value is limited. After logarithmic compression, the coefficient remains at a low level; on the contrary, when the strut shows manifestations of impaired continuity such as jamming, uneven pushing and pulling, and sudden speed changes, the deviation between a single moment and the mean state is sharply amplified, resulting in an increase in the absolute difference and further enhancement under the exponential operation, ultimately causing the to increase significantly. Therefore, the higher the value, the more severe the disturbance and the more serious the damage to the continuity of the action during the pushing and pulling process, and thus it can be used as an important quantitative basis for measuring the degree of impairment of the continuity of the pushing and pulling action.

[0033] In this embodiment, the acquisition logic of the mode deviation index is as follows: Extract the dynamic execution deviation information from the preprocessed whole-process characteristic information, specifically including the actual acceleration of the automotive tailgate strut and the pushing and pulling resistance received by the automotive tailgate strut at different moments within a period of time when the phenomenon of impaired continuity of the pushing and pulling action occurs, and respectively calibrate them as and , represents the actual acceleration of the vehicle tailgate strut at a certain moment within a period when the phenomenon of impaired coherence of the push-pull action occurs, and represents the push-pull resistance received by the vehicle tailgate strut at a certain moment within a period when the phenomenon of impaired coherence of the push-pull action occurs, where , is a positive integer; During the push-pull action detection process, in order to achieve real-time acquisition of the actual acceleration and push-pull resistance of the vehicle tailgate strut, the system integrates an acceleration sensor and a push-pull force sensor on the detection actuator, constructs a data acquisition channel, and transmits various sensing signals to the central processing module in real time. The acceleration sensor can be installed on the structural member near the strut connection point to sense the linear acceleration change of the strut during the telescopic process. The collected signals will be sampled at a fixed time interval to generate an acceleration sequence. The system calibrates the acceleration value at each moment as in the software according to the sampling number, which is used to characterize the dynamic response intensity of the action at that moment. At the same time, the push-pull force sensor is installed on the driving end or the intermediate force transmission component that applies the driving force, and is used to sense the magnitude of the pulling force or pressure applied to the strut by the system in real time. It outputs a digital signal through the force-electricity conversion module and is recorded as the push-pull resistance value by the upper computer, representing the force state of the strut at that moment. After the data acquisition is completed, the software system will automatically perform synchronous processing of the time axis, data denoising, and unit normalization operations to ensure the accurate one-to-one correspondence between the acceleration and the push-pull resistance. The actual acceleration is used to reveal the inertial response characteristics of the strut structure during the action, and to reflect whether there are hysteresis, mutation, or unstable behaviors; while the push-pull resistance is used to reflect the load situation and friction state changes during the action, and is an important physical index for identifying factors such as insufficient lubrication, abnormal sealing, or component wear that cause impaired coherence. These two types of data together constitute the multi-dimensional monitoring basis for action stability and structural integrity, and provide the core input basis for the calculation of the subsequent pattern deviation index.

[0034] By performing cluster analysis and statistical modeling on the acceleration and push-pull resistance data of historical qualified strut samples at each moment, the standard acceleration reference value and the standard push-pull resistance reference value are determined, and are respectively calibrated as and ; To determine the standard acceleration reference value and the standard push-pull resistance reference value for comparison, the system archives the acceleration and push-pull resistance data at different sampling times within a certain period collected during the detection of historical qualified strut samples based on a software platform, and organizes them structurally along the time axis to form a two-dimensional behavior data matrix across samples. Subsequently, at each standard time point, the system takes the acceleration and push-pull resistance data of all samples at that moment as clustering objects, and automatically classifies the data points with similar behavior patterns by calling the K-means algorithm or density-based clustering algorithms (such as DBSCAN). After removing the deviated samples, the central value of the main cluster is extracted as the standard reference value at the current moment. To improve robustness, after clustering, the system also performs Gaussian fitting or kernel density estimation on the sample distribution at each time point to further smooth the change trend of each reference value and avoid mutations caused by individual sample anomalies or detection fluctuations. The entire process is automatically completed by a software module without relying on manual intervention and supports dynamic updating of the model during the accumulation of new samples. Finally, the system will form a set of stable and reusable reference template data sequences at all standard times, corresponding to the standard acceleration reference value and the standard push-pull resistance reference value respectively, as the basis for comparing the deviation degree of the current strut action mode in the follow-up.

[0035] Calculate the mode deviation index, and the specific calculation formula is as follows: In the formula, is the mode deviation index.

[0036] To accurately evaluate the overall deviation degree between the push-pull action process of the vehicle tailgate strut and the standard action template, the calculation formula of the mode deviation index adopts a cumulative summation method based on the absolute value difference of acceleration and the non-linear amplification difference of push-pull resistance. In this formula, first, the actual acceleration of the vehicle tailgate strut at each moment is processed by taking the square root of the square, that is , to ensure that the acceleration uniformly reflects the action intensity regardless of the positive or negative direction, avoiding directional interference. At the same time, by taking the absolute value difference with the standard acceleration reference value , the deviation degree between the dynamic characteristics at the current moment and the normal state is quantified; secondly, the exponential operation is performed on the actual push-pull resistance at each moment to non-linearly amplify the abnormal performance caused by minute resistance fluctuations, and the absolute value difference is calculated with the standard push-pull resistance reference value to enhance the sensitivity to subtle mechanical anomalies; subsequently, the acceleration difference and the push-pull resistance difference at each moment are summed to comprehensively reflect the overall deviation amplitude at that moment. Finally, the deviation values at all moments are averaged to ensure that the evaluation result not only reflects the anomaly at a single moment but also comprehensively reflects the overall trend of the degradation of action coherence over a period of time. Calculated in this way, The higher the value, the more obvious the deviation of the pushing and pulling action behavior pattern from the standard template, the worse the stability of the pole action, and the more serious the degree of coherence impairment.

[0037] Pattern deviation index The magnitude of the value of is positively correlated with the degree of impairment of the pushing and pulling action coherence impairment phenomenon, and its essence reflects the overall deviation degree between the dynamic behavior characteristics of the vehicle tailgate strut during pushing and pulling and the standard action template. When the strut action is stable and the structural performance is normal, the acceleration and pushing and pulling resistance curves at each moment are highly consistent with the standard template. The difference terms generated in the calculation are small, and the final index value is at a low level; while when abnormal fluctuations, sudden increase in friction, mechanical unevenness or response lag occur during the pushing and pulling of the strut, the actual performance of the acceleration and resistance will deviate continuously from the template, and the difference terms will accumulate continuously, ultimately causing the index to increase significantly. Therefore, The larger the value, the more serious the deviation of the pushing and pulling action from the normal state in the overall behavior pattern. The system can quantify the severity of the pushing and pulling action coherence impairment based on this, and it is one of the important bases for determining the impairment level.

[0038] In this embodiment, for the generated action fluctuation coefficient and the pattern deviation index an impairment degree evaluation model is constructed, and the impairment coefficient is generated by weighted summation. The specific calculation formula is as follows: In the formula, is the impairment coefficient, and are the non-zero weight coefficients of the action fluctuation coefficient and the pattern deviation index respectively, and .

[0039] In order to realize the fusion evaluation of the action fluctuation coefficient and the pattern deviation index , the system constructs an impairment degree evaluation model at the software level, performs weighted summation according to the preset weight coefficients, and generates a comprehensive impairment coefficient . This process is automatically completed by the model calculation module. Among them, after the system completes the calculation of and , the two are input into the fusion model in floating-point form and processed according to the formula of . The weight coefficients and are both non-zero real numbers, which are used to regulate and The contribution ratio in the final damage assessment, the sum of the two is fixed at 1 to ensure that the calculation results are maintained under a unified dimension. The weight setting can be based on and The discrimination effect on the actual fault situation is automatically optimized by methods such as the minimum error method, regression fitting, or machine learning models. For example, when historical analysis shows that is more sensitive to the degree of damage than , the system can dynamically set a larger value to increase the recognition weight of the behavior pattern difference. The finally generated damage coefficient comprehensively considers the performance of the strut in two dimensions: action stability and behavior deviation, and can be used to more finely divide the damage level and drive subsequent processing strategies.

[0040] In this embodiment, a preset damage coefficient threshold interval is determined, and after determination, it is compared with the generated damage coefficient to evaluate the degree of damage when the phenomenon of continuous damage in the push-pull action occurs. The specific comparison and analysis are as follows: If , the degree of damage when the phenomenon of continuous damage in the push-pull action occurs is low; This situation indicates that the deviation between the dynamic behavior parameters of the strut and the standard template during the push-pull action is extremely small, that is, the action fluctuation is weak, the behavior pattern remains stable, and there are no obvious non-continuous, stuck, or friction abnormal situations. In this case, the strut shows good structural integrity and action coherence during the test cycle, belonging to a typical normal sample, indicating that its performance has not been significantly affected by wear or aging. The system judges its degree of damage to be low, which means that the strut can be directly classified as a qualified product and enter the subsequent assembly process without re-inspection or special treatment, thus improving the detection efficiency and reducing costs.

[0041] If , the degree of damage when the phenomenon of continuous damage in the push-pull action occurs is medium; This situation indicates that the strut has shown a certain degree of volatility or action pattern deviation during the detection cycle, which may be manifested as occasional jerks during the push-pull process, slightly discontinuous acceleration, or abnormal force peaks. Although it has not reached the level of severe failure, its behavior trend has deviated from the standard template, and there are initial signs of potential structural looseness, lubrication degradation, or component wear. The system determines that it is damaged to a medium degree, which means that although the strut has not shown obvious faults for the time being, there are potential risks in its stability. Usually, it needs to be transferred to the re-inspection or secondary action test link for confirmation to prevent potential quality problems from flowing into the general assembly link and ensure the consistency and reliability of the finished product.

[0042] If , when there is damage to the continuity of the pushing and pulling actions, the degree of damage is the severity level.

[0043] This situation indicates that the strut has experienced a significant degradation in action continuity during the test. Its action fluctuates violently, and the changes in acceleration and resistance deviate significantly from the standard template, possibly accompanied by obvious jamming, abnormal friction, or movement interruption. This state is usually caused by lubricant failure, component fatigue damage, or assembly deviation and has reached an unacceptable failure boundary. The system determines that its degree of damage is severe, meaning that the strut no longer meets the usage requirements and must be removed from the production process as a non-conforming product to prevent the risk of the tailgate not opening and closing smoothly, early failures, or triggering user complaints and large-scale warranty claims after subsequent installation.

[0044] According to the evaluation results of the degree of damage, corresponding measures are implemented respectively; In this embodiment, according to the evaluation results of the degree of damage, corresponding measures are implemented respectively, specifically: If the evaluation result is that the degree of damage is low, the specific measure to be implemented is: directly determine the corresponding strut as qualified and allow it to enter the subsequent assembly process; When the evaluation result is that the degree of damage is low, the software triggers the low-degree status response process. This process automatically identifies the "low-degree" result label through the comparison logic module in the background, writes the qualified identification to the data management module, and at the same time binds the unique code of the strut to the test result and transmits it to the MES (Manufacturing Execution System) interface to achieve status synchronization on the production line. Once the qualified status is confirmed, the system includes the strut in the "automatic release queue", and the control logic drives the mechanical platform to convey it to the subsequent assembly unit, omitting manual review or buffer zone diversion, thus simplifying the process and improving the cycle efficiency. This method ensures automatic determination and release without sacrificing quality, helps shorten the response cycle in the inspection-assembly chain, and improves the circulation rate of qualified products.

[0045] If the evaluation result is that the degree of damage is medium, the specific measure to be implemented is: perform a re-inspection process on the corresponding strut, conduct repeated tests and supplementary tests on the action characteristics, and re-classify and process according to the re-inspection results; When the evaluation result shows that the damage degree is medium, the system will automatically mark the strut as "medium degree" and trigger a re-inspection process through the workflow engine. In the specific operation, the software writes this result into the exception handling queue, calls the re-inspection scheduling logic, designates the strut as a "priority object for re-inspection", guides it to be sent into the secondary action characteristic detection module, repeats the push-pull action several times, and simultaneously collects new full-process characteristic information. The re-inspection data will be compared through a difference evaluation model with the initial data. If the difference decreases significantly or the status is stable, the strut can be re-classified as qualified; otherwise, it will be changed to an unqualified status and the rejection process will be executed. This strategy allows the system to tolerate minor random disturbances or marginal anomalies, improves the fault tolerance of the detection system to minor fluctuations, while preventing the loss of qualified products caused by misjudgment, and ensures the balance between the good product rate and risk control.

[0046] If the evaluation result shows that the damage degree is severe, the specific measures to be taken are: judging the corresponding strut as a defective product and removing it from the assembly process and sending it to the defective product handling station for isolation.

[0047] When the evaluation result shows that the damage degree is severe, the software will immediately mark the strut as "severe degree" and trigger the defective product rejection logic. At this time, the system writes the identification code and defective grade information of the strut into the quality database, and at the same time controls the mechanical sorting device through industrial control instructions to divert the strut from the main line conveying path to the defective product isolation area. To ensure the consistency of subsequent processing, the system will also upload the batch number, detection data, and judgment reason to the central quality traceability module and generate an alarm prompt in the interface system for quality inspection personnel to review. This operation ensures that obviously abnormal struts will not flow into the subsequent links, blocking potential quality hazards from the source, which is a key mechanism to ensure product stability and customer satisfaction, fully realized by software automation, with high responsiveness and full-process traceability.

[0048] Store the detection data, evaluation results, and processing records in the database, and continuously optimize the detection strategy and quality trend analysis based on historical data.

[0049] After the system completes the detection of the push-pull action of the tailgate strut, the assessment of the damage degree, and the processing decision, all the core data related to this process will be structurally encapsulated by the data management module and written into the backend database. The stored content includes, but is not limited to, the characteristic data of the whole process (such as time series of displacement, speed, acceleration, resistance, etc.), the calculated intermediate parameters (such as action fluctuation coefficient, mode deviation index, damage coefficient), the determined damage level by the system, and the finally executed processing measures (such as release, reinspection, rejection, etc.). After each processing is completed, the software system automatically calls the database interface through the data writing instruction, stores the above information in a structured format (such as JSON, CSV, or SQL table) in the quality database, and indexes it according to fields such as strut code, detection time, and processing result, realizing the retrievability, traceability, and multi-dimensional query ability of the data. This mechanism ensures that a complete historical record chain is formed for each detected object, providing basic data support for subsequent data analysis and quality management.

[0050] To achieve the continuous optimization of the detection strategy and the long-term analysis of the product quality trend, the system regularly conducts batch statistics and model operations on the stored detection historical data through the data analysis module built into the software platform. In the specific implementation process, the system can screen historical samples according to dimensions such as time period, batch number, abnormal category, or parameter fluctuation range, and use clustering analysis, regression models, or machine learning algorithms to identify the typical patterns, boundary characteristics, and evolution trends of abnormal occurrences. For example, by comparing the distribution differences of the initial characteristic data of struts with different damage levels, the system can dynamically adjust the weights of the action fluctuation coefficient and the mode deviation index, and optimize the assessment model of the damage coefficient; another example is that by observing the decreasing trend of the pass rate in a certain time period, the system can trigger an automatic alarm and suggest adjusting the detection parameters or the assembly process. This software solution based on the data closed-loop feedback mechanism can realize the intelligent evolution of the detection logic and the early warning of quality problems, which is the key supporting ability to ensure the long-term stable operation of the system and the continuous improvement of the good product rate.

[0051] The above formulas are all dimensionless and take their numerical calculations. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters in the formulas are set by technicians in this field according to the actual situation.

[0052] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions according to the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired or wireless (such as infrared, wireless, microwave, etc.) manner. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that contains one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0053] It should be understood that in various embodiments of the present application, the magnitudes of the sequence numbers of the above processes do not mean the order of execution, and the order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0054] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present application.

[0055] In several embodiments provided by the present application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the above-described embodiments are merely illustrative. For example, the division of units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be an indirect coupling or communication connection through some interfaces, devices or units, and can be in an electrical, mechanical or other form.

[0056] The unit described as a separation component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place or may be distributed across multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0057] In addition, each functional unit in various embodiments of the present application may be integrated in a processing unit, may exist physically separately for each unit, or two or more units may be integrated in one unit.

[0058] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A one-stop offline detection method for a car tailgate strut, characterized in that, Specifically, it includes the following steps: During the process of performing telescopic push-pull cyclic operations on the tailgate strut of the vehicle, real-time detection is carried out to determine whether there is a phenomenon of impaired coherence in the push-pull actions; When the phenomenon of impaired coherence in the push-pull actions is detected, the whole-process characteristic information of the strut push-pull actions is collected in real time through the deployed sensor network and preprocessed; Analyze the whole-process characteristic information after preprocessing to evaluate the degree of impairment when there is a phenomenon of impaired coherence in the push-pull actions; According to the evaluation results of the degree of impairment, corresponding measures are respectively implemented; Store the detection data, evaluation results and processing records in the database, and continuously optimize the detection strategy and quality trend analysis based on historical data.

2. The one-stop offline detection method for a car tailgate strut according to claim 1, characterized in that Analyze the whole-process characteristic information after preprocessing to evaluate the degree of impairment when there is a phenomenon of impaired coherence in the push-pull actions. Specifically, it includes the following steps: Extract the motion behavior perturbation information and dynamic execution deviation information from the whole-process characteristic information after preprocessing, and analyze them after extraction to generate an action fluctuation coefficient and a pattern deviation index respectively; Construct an impairment degree evaluation model for the generated action fluctuation coefficient and pattern deviation index, and generate an impairment coefficient through weighted summation; Determine the pre-set impairment coefficient threshold interval, and compare it with the generated impairment coefficient after determination. According to the comparison results, evaluate the degree of impairment when there is a phenomenon of impaired coherence in the push-pull actions.

3. A one-stop offline detection method for a tailgate strut of an automobile according to claim 2, characterized in that The acquisition logic of the action fluctuation coefficient is as follows: Extract the motion behavior perturbation information from the preprocessed whole-process characteristic information, specifically including the actual displacement of the vehicle tailgate strut at different moments within a period of time when the phenomenon of impaired coherence of the push-pull action occurs and the actual push-pull speed acting on the vehicle tailgate strut, and respectively calibrate them as and , represents the actual displacement of the vehicle tailgate strut at the moment of within a period of time when the phenomenon of impaired coherence of the push-pull action occurs, represents the actual push-pull speed acting on the vehicle tailgate strut at the moment of within a period of time when the phenomenon of impaired coherence of the push-pull action occurs, , is a positive integer; Calculate the average value of the actual displacement of the tailgate strut of the vehicle at all moments within a period of time when the phenomenon of impaired coherence of the push-pull action occurs , according to the formula: ; Calculate the average value of the actual push and pull speeds acting on the tailgate strut at all times within a period when the phenomenon of impaired continuity of push and pull actions occurs , according to the formula: ; Calculate the action fluctuation coefficient. The specific calculation formula is as follows: In the formula, is the action fluctuation coefficient.

4. The one-stop offline detection method for a tailgate strut of an automobile according to claim 3, characterized in that, The acquisition logic of the pattern deviation index is as follows: Extract dynamic execution deviation information from the preprocessed full-process characteristic information, specifically including the actual acceleration of the vehicle tailgate strut and the push-pull resistance received by the vehicle tailgate strut at different moments within a period of time when the phenomenon of impaired coherence of the push-pull action occurs, and calibrate them respectively as and , represents the actual acceleration of the vehicle tailgate strut at the moment of within a period of time when the phenomenon of impaired coherence of the push-pull action occurs, represents the push-pull resistance received by the vehicle tailgate strut at the moment of within a period of time when the phenomenon of impaired coherence of the push-pull action occurs, , is a positive integer; By performing cluster analysis and statistical modeling on the acceleration and push-pull resistance data of historical qualified pole samples at each moment, the standard acceleration reference value and the standard push-pull resistance reference value are determined and calibrated as and ; Calculate the mode deviation index, and the specific calculation formula is as follows: In the formula, is the mode deviation index.

5. The one-stop offline detection method for a tailgate strut of an automobile according to claim 4, wherein For the generated action fluctuation coefficient and the pattern deviation index Construct a damage degree evaluation model, and generate a damage coefficient through weighted summation. The specific calculation formula is as follows: In the formula, is the damage coefficient, and are the non-zero weight coefficients of the action fluctuation coefficient and the pattern deviation index respectively, and .

6. The one-stop offline detection method for a tailgate strut of an automobile according to claim 5, characterized in that Determine the pre-set damaged coefficient threshold range , and after determination, compare it with the generated damaged coefficient for comparison. Evaluate the degree of damage when the phenomenon of continuous damage to the push-pull action occurs according to the comparison result. The specific comparison and analysis are as follows: If , the degree of impairment when the smoothness of the push-pull action is impaired is low; If , when there is an impairment in the continuity of the pushing and pulling actions, the degree of impairment is moderate; If , when there is an impairment in the coherence of the push-pull action, the degree of impairment is the severity level.

7. A one-stop offline detection method for a tailgate strut of an automobile according to claim 6, characterized in that According to the evaluation results of the degree of impairment, corresponding measures are respectively implemented. Specifically: If the evaluation result shows that the degree of impairment is low, the specific measure to be implemented is: directly determine the corresponding strut as qualified and allow it to enter the subsequent assembly process; If the evaluation result shows that the degree of impairment is medium, the specific measure to be implemented is: perform a re-inspection process on the corresponding strut, conduct repeated detections and supplementary measurements of action characteristics, and re-classify and process according to the re-inspection results; If the evaluation result shows that the degree of impairment is severe, the specific measure to be implemented is: determine the corresponding strut as a non-conforming product and remove it from the assembly process, and send it to the non-conforming product processing station for isolation.

Images