Sealing performance detection method and system for gas spring

Abstract

The invention discloses a sealing performance detection method and system for a gas spring, and relates to the technical field of automatic detection, and the method comprises the steps: obtaining the working state information of the gas spring, analyzing the aging coefficient of a part, adjusting an oil film thickness characteristic threshold value, carrying out the multi-angle image collection at regular intervals, and recognizing the oil film thickness on the surface of a piston rod; generating an oil film thickness distribution diagram; according to the adaptive oil film thickness characteristic threshold value, an abnormal gas spring is judged and recognized; and finally, performing continuous pressure monitoring on the abnormal gas spring by adopting a pressure drop method, and outputting a sealing performance detection result. According to the method, the technical problems of low efficiency, high detection blindness and high misjudgment rate of a traditional gas spring sealing performance detection method in the prior art are solved, and the technical effects of realizing efficient and accurate detection of the gas spring sealing performance, remarkably improving the detection efficiency and reducing the misjudgment rate through combination of high-precision image recognition and a pressure drop method are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of automated detection, and particularly to a method and system for detecting the sealing performance of gas springs. Background Art

[0002] Gas springs are widely used in fields such as automobiles and industrial equipment. As important shock-absorbing components, their sealing performance is crucial for the stability and lifespan of products. Traditional methods for detecting the sealing performance of gas springs mainly rely on physical experiments, such as the pressure drop method or the bubble method, which determine the sealing performance by detecting the pressure change inside the gas spring. However, these methods usually face problems such as low detection efficiency, long time consumption, and inaccurate detection results. Especially in large-scale automated production lines, it is difficult to meet the requirements of high-efficiency and accurate quality control.

[0003] In addition, the oil film thickness has an important impact on the sealing performance of gas springs. In the existing technology, the monitoring of the oil film thickness mostly relies on manual detection or simple sensors, lacking high-precision and automated detection means. Summary of the Invention

[0004] This application provides a method and system for detecting the sealing performance of gas springs, aiming to solve the technical problems of low efficiency, high detection blindness, and high false judgment rate in the traditional gas spring sealing performance detection methods in the existing technology.

[0005] In the first aspect of this application, a method for detecting the sealing performance of gas springs is provided. The method includes: obtaining a plurality of working state information of a plurality of gas springs on an automated assembly production line, analyzing and determining a plurality of component aging coefficients according to the plurality of working state information, compensating and adjusting the standard oil film thickness characteristic threshold, and determining a plurality of adapted oil film thickness characteristic thresholds; regularly collecting multi-angle images of the plurality of gas springs, identifying the oil film thickness information on the surface of the gas spring piston rod according to the image collection results, generating a plurality of oil film thickness distribution maps, and analyzing and determining a plurality of oil film thickness data; respectively performing mapping judgment on the plurality of oil film thickness data according to the plurality of adapted oil film thickness characteristic thresholds to determine a plurality of abnormal gas springs that do not meet the adapted oil film thickness characteristic thresholds; continuously monitoring the pressure of the plurality of abnormal gas springs by using the pressure drop method, and outputting the gas spring sealing performance detection result.

[0006] In the second aspect of the present application, a sealing performance detection system for gas springs is provided. The system includes: an oil film thickness characteristic compensation module, which is used to obtain a plurality of working state information of a plurality of gas springs on an automated assembly line, analyze and determine a plurality of component aging coefficients according to the plurality of working state information, compensate and adjust the standard oil film thickness characteristic threshold, and determine a plurality of adapted oil film thickness characteristic thresholds; an oil film thickness distribution recognition module, which is used to regularly collect multi-angle images of the plurality of gas springs, identify the oil film thickness information on the surface of the piston rod of the gas spring according to the image collection results, generate a plurality of oil film thickness distribution maps, and analyze and determine a plurality of oil film thickness data; an abnormal gas spring recognition module, which is used to perform mapping judgment on the plurality of oil film thickness data respectively according to the plurality of adapted oil film thickness characteristic thresholds to determine a plurality of abnormal gas springs that do not meet the adapted oil film thickness characteristic thresholds; a sealing performance detection module, which is used to continuously monitor the pressure of the plurality of abnormal gas springs respectively by using the pressure drop method and output the gas spring sealing performance detection result.

[0007] One or more technical solutions provided in the present application have at least the following technical effects or advantages: The sealing performance detection method and system for gas springs provided by the present application relate to the technical field of automated detection. By analyzing the working state information of gas springs, the oil film thickness threshold is adjusted, and the oil film thickness is identified by multi-angle image collection to generate a distribution map. According to the adjusted threshold, abnormal gas springs are identified, and the pressure drop method is used to monitor the pressure of the abnormal gas springs, and the sealing performance detection result is output, so as to realize efficient and accurate detection of the sealing performance of gas springs, solve the technical problems of low efficiency, high detection blindness and high misjudgment rate in the traditional gas spring sealing performance detection method in the prior art, and achieve the technical effect of realizing efficient and accurate detection of the sealing performance of gas springs by combining high-precision image recognition and the pressure drop method, significantly improving the detection efficiency and reducing the misjudgment rate. Description of the Drawings

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0009] Figure 1 It is a schematic flow chart of the sealing performance detection method for gas springs provided by the embodiments of the present application; Figure 2 It is a schematic structural diagram of the sealing performance detection system for gas springs provided by the embodiments of the present application.

[0010] Explanation of reference numerals: oil film thickness characteristic compensation module 11 , oil film thickness distribution identification module 12 , abnormal gas spring identification module 13 , sealing performance detection module 14 . DETAILED DESCRIPTION

[0011] The present application provides a sealing performance detection method and system for a gas spring, which are used to solve the technical problems in the prior art of low efficiency, high detection blindness and high misjudgment rate of traditional gas spring sealing performance detection methods.

[0012] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0013] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices.

[0014] Embodiment 1, as Figure 1 As shown, the present application provides a method for detecting the sealing performance of a gas spring, the method comprising: P10: Obtain several working status information of several gas springs on the automated assembly production line, determine several component aging coefficients based on the several working status information, make compensation adjustments to the standard oil film thickness characteristic thresholds, and determine several adaptive oil film thickness characteristic thresholds.

[0015] Furthermore, step P10 of the embodiment of the present application further includes: P11: Obtain the working state information of several gas springs, where the working state information at least includes the average working load and the working duration; P12: Quantitatively evaluate the aging state of components according to the average working load and the working duration, and determine several component aging coefficients; P13: Calculate and obtain several adjustment weights according to the several component aging coefficients. If the component aging coefficient is less than or equal to the standard aging coefficient, the adjustment weight is 1. If the component aging coefficient is greater than the standard aging coefficient, calculate the deviation ratio between the component aging coefficient and the standard aging coefficient, and use the sum of the deviation ratio and 1 as the adjustment weight; P14: After compensating and adjusting the standard oil film thickness characteristic threshold by using the several adjustment weights, output several adapted oil film thickness characteristic thresholds.

[0016] It should be understood that the working state information of multiple gas springs on the automated assembly line is obtained, and the aging degree of the sealing components of the gas springs is calculated based on this information, so as to adjust the standard oil film thickness characteristic threshold, and finally generate an adapted oil film thickness characteristic threshold that better conforms to the actual working conditions. This process can dynamically optimize the detection threshold with the operating state of the gas spring, improving the accuracy and adaptability of detection.

[0017] First, obtain the working state information of multiple gas springs from the automated assembly line. The working state information at least includes the average working load and the working duration. The average working load refers to the average load borne by the gas spring during operation, usually in Newtons (N). This parameter can reflect the long-term pressure change trend on the seal and is directly related to its wear degree. The working duration refers to the cumulative operating time of the gas spring, usually in hours (h). A longer working duration may mean that the seal has experienced more severe fatigue aging. To ensure the accuracy of the data, this information can be monitored in real time through industrial Internet of Things sensors and stored and processed by a data acquisition and analysis module.

[0018] Next, evaluate the aging state of the components of the gas spring according to the working state information and calculate the component aging coefficient. This coefficient is used to quantify the aging degree of the seal to provide a basis for adjusting the oil film thickness characteristic threshold in the subsequent process. The calculation method of the component aging coefficient can comprehensively consider the average working load and the working duration of the gas spring. The calculation logic is: when the load borne by the gas spring is high or the operating time is long, the aging degree of its seal increases accordingly, so the value of the aging coefficient will increase. Specifically, by setting a standard reference load and a standard reference duration, compare the average working load and the working duration of the current gas spring with these standard values, and calculate the aging coefficient according to a preset weight coefficient, so as to obtain the aging state of the gas spring. Exemplarily, the calculation formula can be: ; where represents the component aging coefficient, represents the average working load of the current gas spring, represents the standard reference load, represents the working duration of the current gas spring, represents the standard reference duration, , represents the weight coefficient, which is used to balance the effects of load and time on aging. When the value is larger, it indicates that the sealing components of the gas spring are aging more severely, and it may be necessary to adjust the threshold of the oil film thickness characteristic to prevent misjudgment due to the decline in sealing performance.

[0019] Furthermore, the adjustment weight is calculated based on the component aging coefficient for subsequent compensation of the oil film thickness characteristic threshold. When the component aging coefficient is less than or equal to the standard aging coefficient, it indicates that the gas spring is within the normal operating range, and the aging of its seals does not exceed the standard value. Therefore, there is no need to adjust the oil film thickness characteristic threshold, and the adjustment weight is set to 1 at this time. When the component aging coefficient is greater than the standard aging coefficient, it indicates that the aging degree of the seals has exceeded the standard range, and it is necessary to compensate and adjust the oil film thickness characteristic threshold to ensure that the detection standard adapts to gas springs with different aging degrees. In this case, calculate the deviation ratio of the component aging coefficient relative to the standard aging coefficient, that is, the difference between the aging coefficient and the standard aging coefficient divided by the standard aging coefficient, and add 1 to this deviation ratio as the adjustment weight. In this way, the more severe the aging degree of the gas spring, the larger the adjustment weight, and the greater the subsequent compensation and adjustment amplitude.

[0020] Finally, the standard oil film thickness characteristic threshold is compensated and adjusted using the adjustment weight to generate an adapted oil film thickness characteristic threshold. Specifically, the system multiplies the standard oil film thickness characteristic threshold by the adjustment weight to compensate for the impact of aging factors on the sealing performance and ensure the accuracy of oil film thickness detection. The adjusted threshold can better adapt to the actual operating conditions of the gas spring and avoid misjudgment caused by seal aging. When the sealing components of the gas spring are less aged, the system keeps the original oil film thickness characteristic threshold unchanged to ensure the rigor of the detection standard; when the sealing components of the gas spring are more severely aged, the system appropriately relaxes the oil film thickness characteristic threshold to adapt to the aged sealing state and reduce misjudgment or unnecessary repairs and replacements caused by misjudgment. Finally, the system outputs a set of adapted oil film thickness characteristic thresholds and applies them to the subsequent image recognition detection link to improve the accuracy and adaptability of gas spring sealing performance detection.

[0021] Furthermore, step P14 of the embodiment of the present application further includes: P14-1: Obtain the standard oil film thickness characteristic threshold, where the standard oil film thickness characteristic threshold includes the standard maximum thickness, the standard minimum thickness, and the standard thickness uniformity coefficient; P14-2: Positively adjust the standard maximum thickness using the several adjustment weights, negatively adjust the standard minimum thickness and the standard thickness uniformity coefficient using the several adjustment weights, and output the several adapted oil film thickness characteristic thresholds.

[0022] Optionally, the specific process of compensating and adjusting the standard oil film thickness characteristic threshold using the several adjustment weights can be as follows. First, obtain the standard oil film thickness characteristic threshold, which is used to evaluate the oil film thickness distribution on the surface of the gas spring piston rod to judge its sealing performance. The standard oil film thickness characteristic threshold consists of three key parameters: the standard maximum thickness, the standard minimum thickness (the area ratio is greater than a predetermined ratio), and the standard thickness uniformity coefficient. Among them, the standard maximum thickness represents the maximum allowable thickness of the oil film under normal sealing conditions. An overly thick oil film may indicate excessive accumulation of lubricant or leakage risk; the standard minimum thickness refers to the lowest acceptable thickness of the oil film, but it is only valid when the area ratio of this thickness exceeds a predetermined ratio to prevent local extremely thin areas from misleading the overall judgment; the standard thickness uniformity coefficient is used to evaluate the distribution uniformity of the oil film on the piston rod surface. Poor uniformity may mean poor sealing or abnormal oil film coating. By obtaining these parameters, the system can provide reference data for subsequent compensation and adjustment to ensure that the detection standard can adapt to gas springs in different aging states.

[0023] Furthermore, dynamically adjust the standard oil film thickness characteristic threshold based on the adjustment weights to adapt to different degrees of sealing aging states. The adjustment method includes two aspects: positive adjustment and negative adjustment. First, positively adjust the standard maximum thickness using the adjustment weights, that is, when the aging degree of the sealing component increases, allow the maximum thickness of the oil film to increase appropriately to meet the oil film compensation requirement due to seal wear. Second, negatively adjust the standard minimum thickness and the standard thickness uniformity coefficient, that is, as the sealing component ages, the requirement for the standard minimum thickness is appropriately reduced to prevent misjudging seal failure due to local thinning of the oil film, and at the same time relax the requirement for the thickness uniformity coefficient to enable the detection system to tolerate a certain degree of oil film non-uniformity. This adjustment method can ensure that the evaluation of the gas spring sealing performance not only conforms to the actual use situation but also effectively reduces the misjudgment risk caused by aging. Finally, output a set of adapted oil film thickness characteristic thresholds and apply them to the subsequent oil film thickness detection process to improve the accuracy and adaptability of the sealing performance detection.

[0024] P20: Regularly collect multi-angle images of the several gas springs, identify the oil film thickness information on the surface of the gas spring piston rod according to the image collection results, generate several oil film thickness distribution maps, and analyze and determine several oil film thickness data.

[0025] Furthermore, step P20 of the embodiment of the present application further includes: P21: Stitch the multi-angle images of the plurality of gas springs to obtain a plurality of global images, and analyze and determine a plurality of image precisions and a plurality of interference intensities; P22: Pre-train an oil film thickness recognition plug-in, where the oil film thickness recognition plug-in includes K oil film thickness recognition branches, and K is an integer greater than 10; P23: Evaluate and determine a plurality of image error coefficients according to the plurality of image precisions and the plurality of interference intensities, and set a plurality of branch selection quantities, where the branch selection quantity is obtained by rounding the ratio of the image error coefficient to the historical maximum error coefficient multiplied by K; P24: Randomly select within the K oil film thickness recognition branches according to the plurality of branch selection quantities, perform oil film thickness recognition on the plurality of global images, calculate the mean value to obtain a plurality of oil film thickness information, and generate a plurality of oil film thickness distribution maps.

[0026] Specifically, multi-angle images of a plurality of gas springs are collected regularly, the oil film thickness information on the surface of the gas spring piston rod is recognized by computer vision technology, and an oil film thickness distribution map is generated based on the recognition result, and finally the oil film thickness data is analyzed and extracted. The core of this process is to improve the detection accuracy and ensure that the sealing performance of the gas spring can be accurately reflected under complex lighting conditions and different detection environments.

[0027] To achieve this goal, first stitch the images taken from multiple angles to obtain complete global images, and perform quality analysis on these global images to extract image precision and interference intensity. Among them, image precision is an important indicator to measure the clarity and detail retention of an image. High-precision images can more accurately reflect the oil film thickness information, while low-precision images may lead to misjudgment; interference intensity refers to the degree to which the image is affected by factors such as noise, light changes, and surface reflection. A higher interference intensity may reduce the accuracy of oil film thickness recognition. By obtaining these parameters, the oil film thickness detection algorithm can be further optimized to improve robustness.

[0028] Then, pre-train an oil film thickness recognition plug-in, which consists of K oil film thickness recognition branches, where K is an integer greater than 10. These recognition branches are based on a deep learning model and are trained for different lighting conditions, surface material reflection characteristics, and oil film thickness distribution patterns to improve recognition accuracy. The design of the K recognition branches is to adaptively select the optimal recognition path according to the environment in actual applications to ensure the detection accuracy under different working conditions.

[0029] Next, calculate the image error coefficient based on the image precision and interference intensity, and set the number of branches to be selected. The image error coefficient measures the overall error level of the image quality and is used to adjust the adaptability of the recognition model. The calculation method of the number of branches to be selected is as follows: multiply the ratio of the image error coefficient to the historical maximum error coefficient by K and round it to obtain the final number of selected branches. This can ensure that when the image quality is high, fewer recognition branches are used to reduce the calculation cost, while when the image quality is low, the number of recognition branches is increased to improve the robustness and accuracy of recognition.

[0030] Furthermore, according to the number of branches to be selected calculated in the previous step, randomly select within the K oil film thickness recognition branches, and perform oil film thickness recognition on the global image. Subsequently, calculate the mean of the recognition results of all selected branches to obtain the final oil film thickness information, and generate an oil film thickness distribution map based on this. Through this strategy, the recognition strategy can be dynamically adjusted under different image qualities and interference conditions to ensure the accuracy of the oil film thickness data, providing a reliable basis for subsequent seal performance detection.

[0031] Furthermore, step P22 of the embodiment of the present application further includes: P22-1: According to the historical detection data of the same type of gas spring, collect the global image set of the sample gas spring and the sample oil film thickness information set as the sample data set, and divide it into K equal parts to obtain K training sets; P22-2: Use the global image of the sample gas spring as the input and the sample oil film thickness information as the supervision, and use the K training sets to respectively perform supervised training on the convolutional neural network until the mean square error loss function converges, obtaining K oil film thickness recognition branches, and combining them to construct the oil film thickness recognition plug-in.

[0032] Optionally, the detailed process of pre-training the oil film thickness recognition plug-in can be as follows: First, establish a sample data set according to the historical detection data of the same type of gas spring. This data set includes the global image set of the sample gas spring and the sample oil film thickness information set. The global image set of the sample gas spring is spliced after being taken from multiple angles and can comprehensively display the oil film distribution on the surface of the gas spring piston rod; the sample oil film thickness information set comes from high-precision physical measurements and provides the true oil film thickness data corresponding to each sample as the supervision signal for the deep learning model training. To improve the generalization ability of the training, the sample data set is divided into K equal parts to obtain K training sets, and each training set contains a complete image-oil film thickness information pair for subsequent independent training of multiple oil film thickness recognition branches.

[0033] Further, using a convolutional neural network (CNN), with the global image of the sample gas spring as the input and the sample oil film thickness information as the supervision signal, supervised training is performed on each of the K training sets. Through multiple convolutional and pooling operations, the CNN extracts the oil film features in the image and learns the representation forms of different oil film thicknesses on the image. During the training process, the mean squared error (MSE) loss function is used, and the difference between the predicted oil film thickness value and the true oil film thickness value is used as the optimization objective. The backpropagation and gradient descent algorithms are adopted to continuously adjust the network weight parameters until the loss function converges, that is, the prediction error of the model is stable within the optimal range. After the training is completed, K oil film thickness recognition branches are obtained. Each branch is optimized based on a different training set and can adapt to different detection environments and image qualities. Finally, the K recognition branches are combined to construct an oil film thickness recognition plug-in, so that the optimal recognition path can be flexibly called in practical applications to improve the robustness and accuracy of the oil film thickness detection.

[0034] Further, step P20 of the embodiment of the present application further includes: P25: Randomly select a first oil film thickness distribution map, and extract the maximum and minimum oil film thicknesses greater than a predetermined area ratio in the first oil film thickness distribution map to obtain the first maximum oil film thickness and the first minimum oil film thickness; P26: Calculate the thickness standard deviation and thickness deviation ratio of the first oil film thickness distribution map, and obtain the first thickness uniformity coefficient after weighted calculation; P27: Set the first maximum oil film thickness, the first minimum oil film thickness, and the first thickness uniformity coefficient as the first oil film thickness data and add them to the several oil film thickness data.

[0035] In a possible embodiment of the present application, to analyze and determine several oil film thickness data, a first oil film thickness distribution map is randomly selected, and feature extraction is performed on the image to determine the key parameters of the oil film thickness. First, regions with an area ratio greater than a predetermined threshold (such as 1%) are screened out to avoid affecting the overall evaluation due to local outliers (such as noise and extreme values). Then, the maximum and minimum oil film thickness values are extracted within the screened regions and used as the first maximum oil film thickness and the first minimum oil film thickness respectively. The maximum thickness reflects the peak accumulation of the oil film, and an excessive value may indicate abnormal lubricant distribution or poor local sealing; the minimum thickness measures the lowest distribution level of the oil film, and a too small value may mean insufficient lubrication or a risk of seal leakage. Through this process, it is ensured that the extracted thickness data is representative and not affected by local errors.

[0036] Next, further analyze the thickness uniformity of the first oil film thickness distribution map to evaluate the overall stability of the oil film. First, calculate the thickness standard deviation of this distribution map, which is the degree of dispersion of the oil film thickness values of all pixel points relative to the mean value. The larger the standard deviation, the more uneven the oil film thickness distribution. Subsequently, calculate the thickness deviation ratio, which is the relative change amount between the maximum thickness and the minimum thickness of the first oil film. This ratio is used to measure the fluctuation range of the oil film thickness. Finally, perform a weighted calculation on the thickness standard deviation and the thickness deviation ratio to obtain the first thickness uniformity coefficient. This coefficient is used to quantify the distribution uniformity of the oil film thickness. The smaller the value, the more uniform the oil film distribution and the more stable the sealing state; the larger the value, the more uneven the oil film distribution and the possible risk of local leakage.

[0037] Finally, encapsulate the maximum thickness of the first oil film, the minimum thickness of the first oil film, and the first thickness uniformity coefficient calculated in the previous two steps into the first oil film thickness data, and add it to several oil film thickness data sets to build a complete oil film thickness information database. This information database will be used for subsequent sealing performance evaluation and anomaly detection, so as to ensure that the detection system can accurately identify the sealing state of the gas spring based on comprehensive oil film thickness data.

[0038] P30: Respectively perform mapping judgments on the several oil film thickness data according to the several adapted oil film thickness characteristic thresholds, and determine multiple abnormal gas springs that do not meet the adapted oil film thickness characteristic thresholds.

[0039] Furthermore, step P30 of the embodiment of the present application further includes: P31: Analyze and determine several recognition error ratios according to the several image error coefficients, perform secondary optimization adjustment on the several adapted oil film thickness characteristic thresholds, and obtain several adjusted adapted oil film thickness characteristic thresholds; P32: Respectively perform mapping judgments on the several oil film thickness data according to the several adjusted adapted oil film thickness characteristic thresholds. If the maximum oil film thickness is greater than the standard maximum thickness and / or the minimum oil film thickness is less than the standard minimum thickness and / or the thickness uniformity coefficient is less than the standard thickness uniformity coefficient, then set the corresponding gas spring as an abnormal gas spring and output the multiple abnormal gas springs.

[0040] It should be understood that based on the adapted oil film thickness characteristic thresholds calculated in the early stage, perform mapping judgments on all the collected oil film thickness data to identify abnormal gas springs that do not meet the sealing performance standards. The core of the mapping judgment is to compare the oil film thickness data with the adapted thresholds to determine whether there are abnormal situations. If the oil film thickness data of a certain gas spring exceeds the adapted threshold range, then this gas spring is determined to be abnormal and enters the subsequent precise detection link of the sealing performance.

[0041] First, further consider the error impact during the image acquisition process and perform a secondary optimization adjustment on the adapted oil film thickness feature threshold. Specifically, based on the previously calculated image error coefficient, analyze and determine the recognition error ratio. The recognition error ratio reflects the recognition deviation caused by factors such as light changes, reflection interference, and image resolution. If the recognition error ratio is high, it may lead to deviations in the determination results of some oil film thickness data. Therefore, use this recognition error ratio to compensate and adjust the adapted oil film thickness feature threshold to obtain a new set of adjusted adapted oil film thickness feature thresholds to ensure the accuracy of detection.

[0042] Furthermore, based on the adjusted adapted oil film thickness feature threshold, perform a final mapping judgment on all oil film thickness data. The judgment rules are as follows: If the maximum oil film thickness is greater than the standard maximum thickness, there may be excessive lubrication or abnormal local oil film accumulation, indicating an abnormal sealing state; if the minimum oil film thickness is less than the standard minimum thickness, it means insufficient local lubrication, which may lead to seal leakage; if the thickness uniformity coefficient is less than the standard thickness uniformity coefficient, it means that the oil film distribution is uneven, and there may be problems such as poor sealing or increased wear.

[0043] If any one of the conditions is met, the gas spring is marked as an abnormal gas spring and an abnormal result is output to enter the next pressure drop method seal performance detection. This process takes into account both the aging states of different gas springs and the compensation for image errors, making the determination results more reliable.

[0044] P40: Use the pressure drop method to continuously monitor the pressure of the multiple abnormal gas springs respectively and output the gas spring seal performance detection results.

[0045] Specifically, use the pressure drop method to continuously monitor the pressure of the samples determined to be abnormal gas springs to further evaluate their sealing performance. The pressure drop method is a high-precision leak detection technology, and its core principle is: apply a set initial pressure inside the gas spring and continuously monitor the pressure change within a certain period of time. If the pressure drop rate exceeds a predetermined threshold, it means that the gas spring has a leak.

[0046] Specifically, first, apply a standard air pressure (such as 2.5 MPa or the rated working pressure set according to specific product specifications) to each abnormal gas spring, and form a stable sealed gas environment inside the gas spring. Subsequently, start the high-precision pressure sensor, record the real-time pressure value, and conduct time-series analysis to calculate the pressure drop rate and leakage rate. According to the leakage situation, output the following sealing performance test results: qualified (no leakage or leakage within the allowable range), that is, the pressure drop rate is low, and the leakage rate is less than the set safety threshold (such as 0.05 MPa / min), indicating that the gas spring has good sealing performance and can continue to be put into use. Unqualified (leakage exceeds the allowable range), that is, the pressure drop rate is higher than the set threshold (such as 0.1 MPa / min), indicating that the gas spring has poor sealing performance and there is a leakage risk, and it needs to be repaired or replaced.

[0047] The entire monitoring process can adopt multi-point pressure data fitting to ensure the stability and accuracy of the test results. At the same time, a temperature compensation model can be combined to eliminate the gas expansion or contraction effect caused by environmental temperature changes, and further improve the test accuracy. Finally, all test data is stored and a sealing performance test report is generated, providing a reliable basis for subsequent product quality control.

[0048] In summary, the embodiments of the present application have at least the following technical effects: The present application obtains the working state information of the gas spring, analyzes the component aging coefficient, adjusts the characteristic threshold of the oil film thickness, and regularly performs multi-angle image acquisition to identify the oil film thickness on the surface of the piston rod and generate an oil film thickness distribution map. According to the adapted characteristic threshold of the oil film thickness, abnormal gas springs are judged and identified. Finally, the pressure drop method is used to continuously monitor the pressure of the abnormal gas spring, and the sealing performance test results are output, so as to realize the efficient and accurate detection of the sealing performance of the gas spring.

[0049] It achieves the technical effect of realizing the efficient and accurate detection of the sealing performance of the gas spring by combining high-precision image recognition and the pressure drop method, significantly improving the detection efficiency and reducing the misjudgment rate.

[0050] Embodiment 2, based on the same inventive concept as the sealing performance detection method for gas springs in the foregoing embodiment, as Figure 2 shown, the present application provides a sealing performance detection system for gas springs. The system in the embodiments of the present application and the method embodiments are based on the same inventive concept. Among them, the system includes: An oil film thickness characteristic compensation module 11, which is used to obtain the working state information of several gas springs on an automated assembly line, analyze and determine several component aging coefficients according to the several working state information, compensate and adjust the standard oil film thickness characteristic threshold, and determine several adapted oil film thickness characteristic thresholds.

[0051] The oil film thickness distribution recognition module 12 is configured to regularly collect multi - angle images of the plurality of gas springs, identify the oil film thickness information on the surface of the piston rod of the gas spring according to the image collection results, generate a plurality of oil film thickness distribution diagrams, and analyze and determine a plurality of oil film thickness data.

[0052] The abnormal gas spring recognition module 13 is configured to perform mapping judgment on the plurality of oil film thickness data respectively according to the plurality of adapted oil film thickness characteristic thresholds, and determine a plurality of abnormal gas springs that do not meet the adapted oil film thickness characteristic thresholds.

[0053] The sealing performance detection module 14 is configured to continuously monitor the pressure of the plurality of abnormal gas springs respectively by using the pressure drop method, and output the gas spring sealing performance detection result.

[0054] Further, the oil film thickness characteristic compensation module 11 is further configured to perform the following steps: Obtain a plurality of working state information of the plurality of gas springs, where the working state information at least includes the average working load and the working duration; quantitatively evaluate the component aging state according to the average working load and the working duration, and determine a plurality of component aging coefficients; calculate and obtain a plurality of adjustment weights according to the plurality of component aging coefficients, where if the component aging coefficient is less than or equal to the standard aging coefficient, the adjustment weight is 1, if the component aging coefficient is greater than the standard aging coefficient, calculate the deviation ratio between the component aging coefficient and the standard aging coefficient, and use the sum of the deviation ratio and 1 as the adjustment weight; after compensating and adjusting the standard oil film thickness characteristic threshold by using the plurality of adjustment weights, output a plurality of adapted oil film thickness characteristic thresholds.

[0055] Further, the oil film thickness characteristic compensation module 11 is further configured to perform the following steps: Obtain the standard oil film thickness characteristic threshold, where the standard oil film thickness characteristic threshold includes the standard maximum thickness, the standard minimum thickness, and the standard thickness uniformity coefficient; positively adjust the standard maximum thickness by using the plurality of adjustment weights, and reversely adjust the standard minimum thickness and the standard thickness uniformity coefficient by using the plurality of adjustment weights, and output the plurality of adapted oil film thickness characteristic thresholds.

[0056] Further, the oil film thickness distribution recognition module 12 is further configured to perform the following steps: Stitch the multi - angle images of the plurality of gas springs to obtain a plurality of global images, and analyze and determine a plurality of image precisions and a plurality of interference intensities; pre - train an oil film thickness recognition plug - in, where the oil film thickness recognition plug - in includes K oil film thickness recognition branches, and K is an integer greater than 10; evaluate and determine a plurality of image error coefficients according to the plurality of image precisions and a plurality of interference intensities, and set a plurality of branch selection quantities, where the branch selection quantity is obtained by rounding up the ratio of the image error coefficient to the historical maximum error coefficient multiplied by K; randomly select within the K oil film thickness recognition branches according to the plurality of branch selection quantities, and perform oil film thickness recognition on the plurality of global images. After calculating the mean value, obtain a plurality of oil film thickness information and generate a plurality of oil film thickness distribution maps.

[0057] Further, the oil film thickness distribution recognition module 12 is further configured to perform the following steps: According to the historical detection data of the same type of gas springs, collect a sample gas spring global image set and a sample oil film thickness information set as a sample data set, and equally divide it into K parts to obtain K training sets; use the sample gas spring global images as inputs and the sample oil film thickness information as supervision, and use the K training sets to perform supervised training on the convolutional neural network respectively until the mean square error loss function converges, to obtain K oil film thickness recognition branches, and combine and construct the oil film thickness recognition plug - in.

[0058] Further, the oil film thickness distribution recognition module 12 is further configured to perform the following steps: Randomly select a first oil film thickness distribution map, and extract the maximum oil film thickness and the minimum oil film thickness greater than a predetermined area ratio in the first oil film thickness distribution map to obtain a first maximum oil film thickness and a first minimum oil film thickness; calculate the thickness standard deviation and the thickness deviation ratio of the first oil film thickness distribution map, and obtain a first thickness uniformity coefficient after weighted calculation; set the first maximum oil film thickness, the first minimum oil film thickness and the first thickness uniformity coefficient as first oil film thickness data, and add them to the plurality of oil film thickness data.

[0059] Further, the abnormal gas spring recognition module 13 is further configured to perform the following steps: Analyze and determine a plurality of recognition error ratios according to the plurality of image error coefficients, perform secondary optimization adjustment on the plurality of adapted oil film thickness feature thresholds, and obtain a plurality of adjusted adapted oil film thickness feature thresholds; perform mapping judgment on the plurality of oil film thickness data according to the plurality of adjusted adapted oil film thickness feature thresholds respectively. If the maximum oil film thickness is greater than the standard maximum thickness and / or the minimum oil film thickness is less than the standard minimum thickness and / or the thickness uniformity coefficient is less than the standard thickness uniformity coefficient, then set the corresponding gas spring as an abnormal gas spring and output the plurality of abnormal gas springs.

[0060] It should be noted that the above order of the embodiments of the present application is only for description and does not represent the superiority or inferiority of the embodiments. In addition, the specific embodiments of this specification have been described. Moreover, the processes depicted in the drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0061] The above are only the preferred embodiments of the present application and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the protection scope of the present application.

[0062] This specification and the drawings are only exemplary descriptions of the present application and are considered to have covered any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. A method for detecting the sealing performance of a gas spring, characterized in that, The method includes: Obtaining the working state information of several gas springs on an automated assembly line, analyzing and determining several component aging coefficients based on the several working state information, compensating and adjusting the standard oil film thickness characteristic threshold, and determining several adapted oil film thickness characteristic thresholds; Regularly collecting multi-angle images of the several gas springs, identifying the oil film thickness information on the surface of the gas spring piston rod according to the image collection results, generating several oil film thickness distribution maps, and analyzing and determining several oil film thickness data; Performing mapping judgment on the several oil film thickness data respectively according to the several adapted oil film thickness characteristic thresholds, and determining multiple abnormal gas springs that do not meet the adapted oil film thickness characteristic thresholds; Using the pressure drop method to continuously monitor the pressure of the multiple abnormal gas springs respectively, and outputting the gas spring sealing performance detection result.

2. The sealing performance detection method for a gas spring according to claim 1, characterized in that Analyzing and determining several component aging coefficients based on the several working state information, compensating and adjusting the standard oil film thickness characteristic threshold, and determining several adapted oil film thickness characteristic thresholds, including: Obtaining the working state information of several gas springs, where the working state information at least includes the average working load and the working duration; Quantitatively evaluating the component aging state according to the average working load and the working duration, and determining several component aging coefficients; Calculating and obtaining several adjustment weights according to the several component aging coefficients, where if the component aging coefficient is less than or equal to the standard aging coefficient, the adjustment weight is 1, if the component aging coefficient is greater than the standard aging coefficient, calculating the deviation ratio between the component aging coefficient and the standard aging coefficient, and taking the sum of the deviation ratio and 1 as the adjustment weight; After compensating and adjusting the standard oil film thickness characteristic threshold by using the several adjustment weights, outputting several adapted oil film thickness characteristic thresholds.

3. The sealing performance detection method for the gas spring according to claim 2, characterized in that, After compensating and adjusting the standard oil film thickness characteristic threshold by using the several adjustment weights, outputting several adapted oil film thickness characteristic thresholds, including: Obtaining the standard oil film thickness characteristic threshold, where the standard oil film thickness characteristic threshold includes the standard maximum thickness, the standard minimum thickness, and the standard thickness uniformity coefficient; Positively adjusting the standard maximum thickness by using the several adjustment weights, and negatively adjusting the standard minimum thickness and the standard thickness uniformity coefficient by using the several adjustment weights, and outputting the several adapted oil film thickness characteristic thresholds.

4. The method for detecting the sealing performance of a gas spring according to claim 3, characterized in that, Identifying the oil film thickness information on the surface of the gas spring piston rod according to the image collection results, and generating several oil film thickness distribution maps, including: Stitching the multi-angle images of the several gas springs, obtaining several global images, and analyzing and determining several image precisions and several interference intensities; Pre-training an oil film thickness recognition plug-in, where the oil film thickness recognition plug-in includes K oil film thickness recognition branches, and K is an integer greater than 10; Evaluating and determining several image error coefficients according to the several image precisions and several interference intensities, and setting several branch selection quantities, where the branch selection quantity is obtained by rounding up the ratio of the image error coefficient to the historical maximum error coefficient multiplied by K; According to the selected quantity of the several branches, randomly select within the K oil film thickness recognition branches, and perform oil film thickness recognition on the several global images. After calculating the mean value, several oil film thickness information are obtained, and several oil film thickness distribution maps are generated.

5. The sealing performance detection method for a gas spring according to claim 4, characterized in that, Pre-trained oil film thickness recognition plug-in, including: According to the historical detection data of the same type of gas springs, collect a sample gas spring global image set and a sample oil film thickness information set as a sample data set, and equally divide it into K parts to obtain K training sets; Taking the sample gas spring global image as the input and the sample oil film thickness information as the supervision, use the K training sets to respectively perform supervised training on the convolutional neural network until the mean square error loss function converges, obtain K oil film thickness recognition branches, and combine and construct the oil film thickness recognition plug-in.

6. The sealing performance detection method for a gas spring according to claim 4, characterized in that, Analyze and determine several oil film thickness data, including: Randomly select the first oil film thickness distribution map, and extract the maximum oil film thickness and the minimum oil film thickness greater than the predetermined area ratio in the first oil film thickness distribution map to obtain the first maximum oil film thickness and the first minimum oil film thickness; Calculate the thickness standard deviation and the thickness deviation ratio of the first oil film thickness distribution map, and obtain the first thickness uniformity coefficient after weighted calculation; Set the first maximum oil film thickness, the first minimum oil film thickness and the first thickness uniformity coefficient as the first oil film thickness data, and add them to the several oil film thickness data.

7. The sealing performance detection method for a gas spring according to claim 6, characterized in that, According to the several adapted oil film thickness characteristic thresholds, respectively perform mapping judgment on the several oil film thickness data to determine multiple abnormal gas springs that do not meet the adapted oil film thickness characteristic thresholds, including: Analyze and determine several recognition error ratios according to the several image error coefficients, perform secondary optimization adjustment on the several adapted oil film thickness characteristic thresholds, and obtain several adjusted adapted oil film thickness characteristic thresholds; According to the several adjusted adapted oil film thickness characteristic thresholds, respectively perform mapping judgment on the several oil film thickness data. If the maximum oil film thickness is greater than the standard maximum thickness and / or the minimum oil film thickness is less than the standard minimum thickness and / or the thickness uniformity coefficient is less than the standard thickness uniformity coefficient, then set the corresponding gas spring as an abnormal gas spring and output the multiple abnormal gas springs.

8. A sealing performance detection system for a gas spring, characterized in that The system includes: An oil film thickness characteristic compensation module, which is used to obtain several working state information of several gas springs on the automatic assembly production line, analyze and determine several component aging coefficients according to the several working state information, compensate and adjust the standard oil film thickness characteristic thresholds, and determine several adapted oil film thickness characteristic thresholds; An oil film thickness distribution recognition module, which is used to regularly collect multi-angle images of the several gas springs, recognize the oil film thickness information on the surface of the gas spring piston rod according to the image collection results, generate several oil film thickness distribution maps, and analyze and determine several oil film thickness data; Abnormal gas spring identification module, which is used to respectively perform mapping judgment on the plurality of oil film thickness data according to the plurality of adapted oil film thickness characteristic thresholds, and determine a plurality of abnormal gas springs that do not meet the adapted oil film thickness characteristic thresholds; Sealing performance detection module, which is used to respectively perform continuous pressure monitoring on the plurality of abnormal gas springs by using the pressure drop method, and output the gas spring sealing performance detection result.

Images