Intelligent evaluation system for tensile fatigue of down jacket fabric based on cloud computing

Abstract

The invention discloses an intelligent evaluation system for tensile fatigue of a down jacket fabric based on cloud computing, relates to the technical field of down jacket fabrics, and solves the problems that a fine evaluation logic for the elastic recovery capability of the fabric after cyclic stretching is lacked and the influence of the recovery rate difference of different regions on the overall fatigue performance of the fabric is ignored. Differentiation evaluation logics are respectively designed for an abnormal value section and a normal area through a main analysis center; for the abnormal value section, the recovery rate change after cyclic stretching is emphatically monitored, and the anti-fatigue capability of a fabric weak area is focused; for a non-abnormal area, continuously tracking the maximum value of the recovery rate in a plurality of groups of cycle tests, and comprehensively evaluating the overall tensile fatigue stability of the fabric; and in combination with a quantitative standard of less than or equal to 30 cycle tests and a clear judgment rule of a 20% recovery rate threshold value, an evaluation result is upgraded from experience judgment to data driving, and a quantifiable scientific basis is provided for the durability of the fabric.

Description

Technical Field

[0001] This invention relates to the field of down jacket fabric technology, specifically to a cloud computing-based intelligent assessment system for tensile fatigue of down jacket fabrics. Background Technology

[0002] In the production and application of down jacket fabrics, tensile fatigue performance is a core indicator for measuring their durability, structural stability, and lifespan, directly affecting the maintenance of warmth and shape stability of down jackets during long-term wear and repeated activities. Traditional methods for assessing the tensile fatigue of down jacket fabrics largely rely on manual operation and experience, typically involving cyclic tensile testing using a universal testing machine, combined with manual observation of fabric fracture or recording force changes using simple sensors. These methods have several limitations.

[0003] On the one hand, traditional assessment methods are insufficient in monitoring the microscopic deformation of the fabric surface during stretching, making it difficult to accurately capture early fatigue signals in weak areas inside the fabric (such as yarn interlacing points and fiber defects). This results in assessment results that are easily affected by subjective factors, leading to poor accuracy and consistency.

[0004] On the other hand, manual analysis requires the manual organization and calculation of massive amounts of test data, which is not only inefficient but also makes it difficult to track the dynamic changes of feature points during stretching in real time. This makes it impossible to promptly identify abnormal stretching areas (such as locally excessively deformed sections), potentially leading to misjudgments of high-risk fabrics. Furthermore, traditional systems lack a refined evaluation logic for the elastic recovery ability of fabrics after cyclic stretching, often using the number of fractures or maximum tensile force as a single criterion, ignoring the impact of differences in recovery rates in different areas on the overall fatigue performance of the fabric. This makes it difficult to comprehensively reflect the fabric's fatigue resistance performance in actual wearing scenarios. In addition, weak data storage and sharing capabilities hinder cross-departmental collaborative analysis of test results, restricting the efficiency of fabric R&D optimization and quality control.

[0005] Against this backdrop, there is an urgent need to construct a systematic solution that integrates visual monitoring, intelligent analysis, and precise evaluation. Through automated feature recognition, dynamic anomaly calibration, and scientific cyclic testing logic, this solution can achieve intelligent evaluation of the tensile fatigue performance of down jacket fabrics, providing technical support for improving fabric quality and optimizing processes in the textile industry. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a cloud computing-based intelligent assessment system for the tensile fatigue of down jacket fabrics. This system solves the problem of lacking a refined assessment logic for the elastic recovery ability of fabrics after cyclic stretching, and relying primarily on the number of fractures or the maximum tensile force as a single criterion, thus ignoring the impact of differences in recovery rates in different areas on the overall fatigue performance of the fabric.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a cloud computing-based intelligent assessment system for tensile fatigue of down jacket fabrics, comprising:

[0008] At the machine vision end, visual monitoring is performed on the down jacket fabrics participating in the test;

[0009] At the feature point locking end, feature verification is performed on the first acquired surface image of the down jacket fabric. Based on the gradient features associated with the pixels within the image and the stretching directions associated with both sides, the feature points existing in the surface image are confirmed sequentially. The specific method is as follows:

[0010] The pixel values ​​associated with different pixels within the image of the down jacket fabric surface are identified, and the identified pixel values ​​are labeled as X. i Here, i represents different pixels, and the Sobel algorithm is used to confirm the vertical and gradients associated with a given pixel. Then, based on the vertical and gradients associated with the corresponding pixel, the combined gradient associated with the corresponding pixel is determined. Pixels that satisfy the condition that the comprehensive gradient is greater than or equal to Y1 are labeled as gradient pixels, with Y1 being a preset value; otherwise, no labeling is performed.

[0011] Connect the confirmed consecutive gradient pixels to confirm the gradient contour associated with the corresponding gradient pixels, and record the intersection points associated with different gradient contours as points to be selected.

[0012] After the candidate points in the image are confirmed in sequence, several candidate points located in the same pulling direction are connected to confirm the connection according to the preset pulling direction. From the confirmed connection lines, a set of connection lines is randomly selected as the built-in standard line, and several candidate points associated with the built-in standard line are marked as feature points.

[0013] The anomaly calibration processing unit continuously verifies the features of the down jacket fabric surface image during the testing process, identifies the tensile variation values ​​between adjacent feature points, and confirms outlier segments based on several identified tensile variation values. The specific method is as follows:

[0014] During the tensile test process, the original distance length L1 between the two clamping parts is recorded, and the overall tensile length L2 of the down jacket fabric surface image is recorded. If L2 and L1 satisfy: (L2-L1)÷L1≥30%, then the analysis process is executed.

[0015] Based on the identified feature points, identify the straight-line distance L between adjacent feature points. k Where k represents the connection segment between different feature points, and the confirmed straight-line distances L k Perform comparison and verification: randomly select a set of L kAs intermediate data, and based on the preset range value F1, the search range is confirmed: L k ±F1, where F1 is a preset value, includes other L values ​​within this search range. k As supplementary data to the intermediate data, record the total number G of supplementary data, and then select other L in sequence. k As intermediate data, the associated search range is confirmed sequentially, and then the total number G associated with the corresponding auxiliary data is confirmed synchronously. From the confirmation process of several intermediate data and search ranges, L is... k The process for selecting the auxiliary data is denoted as the selected process. From the total number of different data G associated with the selected process, the intermediate data associated with Gmax is selected and denoted as the standard data. Both the standard data and the auxiliary data belonging to the standard data are then confirmed.

[0016] Identify whether all straight-line distances other than the standard data are marked as supplementary data. If so, it means that there are no outlier segments in the current test process.

[0017] If not, then the other linear distances L that were not classified as supplementary data will be... k As an anomaly distance, the maximum value is selected from the confirmed anomaly distances, and the connection segment associated with the maximum value is recorded as the anomaly segment;

[0018] The main analysis center confirms whether there are outlier segments in the surface image of the down jacket fabric. If they exist, the center identifies whether the recovery rate of the outlier segments meets the standard in subsequent cyclic tests and displays the signal based on the identification results. If they do not exist, the center performs feature verification and evaluation on the adjacent segments between multiple adjacent feature points.

[0019] The preferred detailed handling method for segments containing outliers is as follows:

[0020] In subsequent cyclic tests, after the overall stretching length exceeds 30%, the tension is removed, allowing the down jacket fabric to recover by its internal elasticity. The recovery length Lq between the feature points on both sides of the corresponding outlier segment is recorded, where q represents the number of subsequent cyclic tests, and q≤30. The length data Lb between the corresponding feature points in the image of the down jacket fabric surface before the first stretch is then recorded, and the confirmed Lb is recorded as the standard length.

[0021] The evaluation value PDq associated with the corresponding cycle test process is confirmed by (Lq-Lb)÷Lb=PDq. If PDq>20%, a tensile test failure signal is directly generated and displayed through the display unit. If PDq≤20%, the confirmation continues. If none of the confirmed evaluation values ​​PDq exceed 20%, a tensile test compliance signal is directly generated and displayed through the display unit.

[0022] The preferred detailed processing method for segments without outliers is as follows:

[0023] In subsequent cyclic tests, after the overall stretch length exceeded 30%, the tension was released, allowing the down jacket fabric to recover its elasticity. The recovery length between adjacent feature points was confirmed, as well as the original length associated with those adjacent feature points before the initial stretch test. The formula was: (Recovered length - Original length) ÷ Original length = BZ k Confirm the restoration rate BZ between corresponding adjacent feature points k Where k represents the connection segment between different feature points, and BZ represents several sets of recovery rates associated with different connection segments in the corresponding test process. k Select the maximum value BZ k max, if BZ k If the maximum deviation exceeds 20%, a tensile test failure signal is generated and displayed through the display unit. Otherwise, the tensile test continues, and the presence of BZ (Boundary Error) is assessed in subsequent tensile test processes. k Processes with a maximum percentage greater than 20%:

[0024] If present, a tensile test failure signal will be generated directly;

[0025] If it does not exist, continue checking until the cyclical testing process reaches 30 sets, then stop. If none of them exist, BZ will be found. k For processes with a maximum completion rate greater than 20%, a tensile test compliance signal is directly generated and displayed directly through the display unit.

[0026] This invention provides a cloud-based intelligent assessment system for tensile fatigue of down jacket fabrics. Compared with existing technologies, it has the following advantages:

[0027] In terms of anomaly identification efficiency, the anomaly calibration processing end can quickly lock out abnormal value segments (such as weak areas of overstretching) during the stretching process by dynamically defining the standard data range and automatically comparing the distance differences between feature points, thus avoiding the tediousness and omissions of manual inspection. This targeted anomaly focus enables the system to identify potential fabric defects in the early testing stage, providing direction for key monitoring in subsequent cyclic testing and greatly improving the targeting of fatigue assessment.

[0028] From the perspective of the scientific nature of fatigue assessment, the main analysis center designed differentiated assessment logic for outlier and normal areas: for outlier areas, the focus is on monitoring the change in recovery rate after cyclic stretching, highlighting the fatigue resistance of weak areas in the fabric; for areas without outliers, the overall tensile fatigue stability of the fabric is comprehensively assessed by continuously tracking the maximum recovery rate in multiple cyclic tests; combined with the quantitative standard of ≤30 cycles and the clear judgment rule of the 20% recovery rate threshold, the assessment results are upgraded from "experience-based judgment" to "data-driven", providing a quantifiable scientific basis for fabric durability. Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the principle framework of the present invention. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] Please see Figure 1 This application provides a cloud computing-based intelligent assessment system for tensile fatigue of down jacket fabrics, including a machine vision terminal, a feature point locking terminal, an anomaly calibration processing terminal, a main analysis center, and a display unit. The machine vision terminal is electrically connected to the input node of the feature point locking terminal or the main analysis center, and the feature point locking terminal, the anomaly calibration processing terminal, and the main analysis center are electrically connected from the output node to the input node in sequence. The main analysis center is also electrically connected to the input node of the display unit.

[0032] In the machine vision section, visual monitoring is performed on the down jacket fabric participating in the test, and the real-time monitored surface image of the down jacket fabric is transmitted to the feature point locking section. During the test, clamping mechanisms are set on both sides of the fabric to clamp the fabric on both sides. Then, according to the hydraulic components, the fabric is stretched and its internal tensile performance is comprehensively evaluated.

[0033] In the feature point locking stage, the feature verification is performed on the first acquired surface image of the down jacket fabric. Based on the gradient features associated with the pixels within the image and the pulling directions associated with both sides, the feature points existing in the surface image are confirmed sequentially. The specific method for confirming the feature points is as follows:

[0034] The pixel values ​​associated with different pixels within the image of the down jacket fabric surface are identified, and the identified pixel values ​​are labeled as X. iHere, i represents different pixels, and the Sobel algorithm is used to confirm the vertical and gradients associated with a given pixel. Then, based on the vertical and gradients associated with the corresponding pixel, the combined gradient associated with the corresponding pixel is determined. Pixels that satisfy the condition that the comprehensive gradient is greater than or equal to Y1 are labeled as gradient pixels, where Y1 is a preset value, the specific value of which is determined by the operator based on experience; otherwise, no labeling is performed.

[0035] Connect the confirmed continuous gradient pixels to confirm the gradient contour associated with the corresponding gradient pixels, and record the intersection points associated with different gradient contours as the points to be selected. Specifically, gradient pixels will form a set of gradient contours, and there are corresponding intersection points between the gradient contours. The corresponding intersection points are the intersection points of the threads in the fabric. There are obvious protrusions, so the corresponding obvious protrusions are the points to be selected that need to be confirmed.

[0036] After the candidate points within the image are confirmed sequentially, several candidate points located in the same stretching direction are connected according to the preset stretching direction. From the confirmed sets of connections, one set of connections is randomly selected as the built-in standard line. Several candidate points associated with the built-in standard line are marked as feature points. Specifically, after the corresponding candidate points are confirmed, several candidate points are associated with specific connections, and there are associated candidate points between corresponding connections. In the subsequent tensile test process, there will be stretching between the corresponding candidate points. Based on the corresponding stretching characteristics, the stretching verification of the corresponding feature points can be effectively completed, making it easier for the corresponding fabric to achieve a better tensile strength test result.

[0037] The anomaly calibration processing end continuously verifies the features of the down jacket fabric surface image during the test, identifies the tensile variation between adjacent feature points, and confirms the outlier segment based on the identified tensile variation. Specifically, during the tensile test processing, each feature point can be stretched. During stretching, the change values ​​between different feature points may be different or the same. If they are the same, it means there is no outlier segment. If they are different, it means there is an outlier segment, and relevant anomaly judgment is required.

[0038] The specific method for identifying the stretching variation between adjacent feature points is as follows:

[0039] During the tensile test process, the original distance length L1 between the two clamping parts is recorded, and the overall tensile length L2 of the down jacket fabric surface image is recorded. If L2 and L1 satisfy: (L2-L1)÷L1≥30%, then the analysis process is executed (30% is a critical value. When the corresponding length data meets this standard, it means that the corresponding fabric meets the clamping process and the associated tensile strength meets the standard. Then the subsequent analysis process can be analyzed to identify whether there is abnormal behavior in the length segment between the corresponding feature points).

[0040] Based on the identified feature points, identify the straight-line distance L between adjacent feature points. k Where k represents the connection segment between different feature points, and the confirmed straight-line distances L k Perform comparison and verification: randomly select a set of L k As intermediate data, and based on the preset range value F1, the search range is confirmed: L k ±F1, where F1 is a preset value, and its specific value is determined by the operator based on experience. Other L values ​​within this search range will be included. k As supplementary data to the intermediate data, record the total number G of supplementary data, and then select other L in sequence. k As intermediate data, the associated search range is confirmed sequentially, and then the total number G associated with the corresponding auxiliary data is confirmed synchronously. From the confirmation process of several intermediate data and search ranges, L is... k The process of selecting the auxiliary data min (that is, the minimum straight-line distance) is denoted as the selected process. From the total number of different values ​​G associated with the selected process, the intermediate data associated with Gmax (that is, the maximum value of the total number of different values ​​G) is selected and denoted as the standard data. The standard data and the auxiliary data belonging to the standard data are then confirmed.

[0041] Identify whether all straight-line distances other than the standard data are marked as supplementary data. If so, it means that there are no outlier segments in the current test process.

[0042] If not, then the other linear distances L that were not classified as supplementary data will be... k As an anomaly distance, the maximum value is selected from the confirmed anomaly distances, and the connection segment associated with the maximum value is recorded as the anomaly segment;

[0043] Specifically, there are different handling methods for down jacket fabrics with outlier ranges or those without. These methods are implemented by the main analysis center to identify whether the fabric meets the standards in the tensile test during subsequent cyclic testing processes.

[0044] The main analysis center confirms whether there are outlier segments within the image of the down jacket fabric surface. If so, in subsequent cyclic testing, it identifies whether the recovery rate of the outlier segments meets the standard and displays the signal based on the identification results. If not, it performs feature verification and evaluation on adjacent segments between multiple adjacent feature points to determine whether the current down jacket fabric surface image meets the test standards.

[0045] The detailed handling method for segments containing outliers is as follows:

[0046] In subsequent cyclic tests, after the overall stretching length exceeds 30%, the tension is removed, allowing the down jacket fabric to recover by its internal elasticity. The recovery length Lq between the feature points on both sides of the corresponding outlier segment is recorded, where q represents the number of subsequent cyclic tests, and q≤30 (that is, it should not exceed 30 times). The length data Lb between the corresponding feature points in the image of the down jacket fabric surface before the first stretch is also recorded, and the confirmed Lb is recorded as the standard length.

[0047] The evaluation value PDq associated with the corresponding cycle test process is confirmed by (Lq-Lb)÷Lb=PDq. If PDq>20%, a tensile test failure signal is directly generated and displayed through the display unit. Otherwise, the confirmation continues. If none of the confirmed evaluation values ​​PDq exceed 20%, a tensile test compliance signal is directly generated and displayed through the display unit.

[0048] Specifically, during the initial confirmation process, there exists a set of corresponding value segments that are stretched too much. These corresponding value segments are the segments with the smallest elasticity parameter. In the actual processing process, these corresponding value segments are the most difficult to recover. We can directly confirm the process based on the data change process of this segment, thereby comprehensively evaluating the related test compliance and displaying the relevant information to achieve the optimal processing process.

[0049] The detailed handling method for segments without outliers is as follows:

[0050] In subsequent cyclic tests, after the overall stretch length exceeded 30%, the tension was released, allowing the down jacket fabric to recover its elasticity. The recovery length between adjacent feature points was confirmed, as well as the original length associated with those adjacent feature points before the initial stretch test. The formula was: (Recovered length - Original length) ÷ Original length = BZ k Confirm the restoration rate BZ between corresponding adjacent feature points k Where k represents the connection segment between different feature points, and BZ represents several sets of recovery rates associated with different connection segments in the corresponding test process. k Select the maximum value BZ k max, if BZk If the maximum deviation exceeds 20%, a tensile test failure signal is generated and displayed through the display unit. Otherwise, the tensile test continues, and the presence of BZ (Boundary Error) is assessed in subsequent tensile test processes. k If the process rate is greater than 20%, a tensile test failure signal will be generated directly. Otherwise, the process will continue to check until 30 sets of tests are completed. If no BZ signal is found, the process will stop. k For processes where max > 20%, a tensile test pass signal is directly generated and displayed directly through the display unit;

[0051] Specifically, in the subsequent processing, there is a corresponding change in the stretching length. During the continuous testing, there is a change in the recovery rate. As the number of cyclic tests gradually increases, the recovery rate will continue to increase. Based on the subsequent cyclic testing process, it can be determined whether the test meets the standard, and thus a comprehensive display can be made.

[0052] Some of the data in the above formulas are numerical calculations with dimensions removed, and the contents not described in detail in this specification are all prior art known to those skilled in the art.

[0053] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A cloud computing-based intelligent assessment system for tensile fatigue of down jacket fabrics, characterized in that, include: At the machine vision end, visual monitoring is performed on the down jacket fabrics participating in the test; The feature point locking end performs feature verification on the first acquired surface image of the down jacket fabric. Based on the gradient features associated with the pixels inside the image and the pulling direction associated with both sides, the feature points existing in the surface image are confirmed in sequence. The anomaly calibration processing end continuously performs feature verification on the surface image of the down jacket fabric during the test, identifies the tensile variation between adjacent feature points, and confirms the outlier segment based on the identified tensile variation. The main analysis center confirms whether there are outlier segments in the surface image of the down jacket fabric. If they exist, the center identifies whether the recovery rate of the outlier segments meets the standard in subsequent cyclic tests and displays the signal based on the identification results. If they do not exist, the center performs feature verification and evaluation on the adjacent segments between multiple adjacent feature points.

2. The cloud computing-based intelligent assessment system for tensile fatigue of down jacket fabrics according to claim 1, characterized in that, The specific method by which the feature point locking end sequentially confirms the feature points existing in the surface image is as follows: The pixel values ​​associated with different pixels within the image of the down jacket fabric surface are identified, and the identified pixel values ​​are labeled as X. i Here, i represents different pixels, and the Sobel algorithm is used to confirm the vertical and gradients associated with a given pixel. Then, based on the vertical and gradients associated with the corresponding pixel, the combined gradient associated with the corresponding pixel is determined. And the pixels that satisfy the condition: comprehensive gradient ≥ Y1 are marked as gradient pixels, and Y1 is a preset value; Connect the confirmed consecutive gradient pixels to confirm the gradient contour associated with the corresponding gradient pixels, and record the intersection points associated with different gradient contours as points to be selected. After the candidate points in the image are confirmed in sequence, several candidate points located in the same pulling direction are connected to confirm the connection according to the preset pulling direction. From the confirmed connection lines, a set of connection lines is randomly selected as the built-in standard line, and several candidate points associated with the built-in standard line are marked as feature points.

3. The cloud computing-based intelligent assessment system for tensile fatigue of down jacket fabrics according to claim 2, characterized in that, Pixels with a combined gradient < Y1 are not calibrated.

4. The cloud computing-based intelligent assessment system for tensile fatigue of down jacket fabrics according to claim 1, characterized in that, The specific method by which the anomaly calibration processing terminal confirms the outlier value segment is as follows: During the tensile test process, the original distance length L1 between the two clamping parts is recorded, and the overall tensile length L2 of the down jacket fabric surface image is recorded. If L2 and L1 satisfy: (L2-L1)÷L1≥30%, then the analysis process is executed. Based on the identified feature points, identify the straight-line distance L between adjacent feature points. k Where k represents the connection segment between different feature points, and the confirmed straight-line distances L k Perform comparison and verification: randomly select a set of L k As intermediate data, and based on the preset range value F1, the search range is confirmed: L k ±F1, where F1 is a preset value, includes other L values ​​within this search range. k As supplementary data to the intermediate data, record the total number G of supplementary data, and then select other L in sequence. k As intermediate data, the associated search range is confirmed sequentially, and then the total number G associated with the corresponding auxiliary data is confirmed synchronously. From the confirmation process of several intermediate data and search ranges, L is... k The process for selecting the auxiliary data is denoted as the selected process. From the total number of different data G associated with the selected process, the intermediate data associated with Gmax is selected and denoted as the standard data. Both the standard data and the auxiliary data belonging to the standard data are then confirmed. Identify whether all straight-line distances other than the standard data are marked as supplementary data. If so, it means that there are no outlier segments in the current test process. If not, then the other linear distances L that were not classified as supplementary data will be... k As an outlier distance, the maximum value is selected from the confirmed outlier distances, and the connection segment associated with the maximum value is recorded as the outlier segment.

5. The cloud computing-based intelligent assessment system for tensile fatigue of down jacket fabrics according to claim 1, characterized in that, The main analysis center handles outlier segments in detail as follows: In subsequent cyclic tests, after the overall stretching length exceeds 30%, the tension is removed, allowing the down jacket fabric to recover by its internal elasticity. The recovery length Lq between the feature points on both sides of the corresponding outlier segment is recorded, where q represents the number of subsequent cyclic tests, and q≤30. The length data Lb between the corresponding feature points in the image of the down jacket fabric surface before the first stretch is then recorded, and the confirmed Lb is recorded as the standard length. The evaluation value PDq associated with the corresponding cycle test process is confirmed by using (Lq-Lb)÷Lb=PDq. If PDq>20%, a tensile test failure signal is directly generated and displayed through the display unit.

6. The cloud computing-based intelligent assessment system for tensile fatigue of down jacket fabrics according to claim 5, characterized in that, If PDq≤20%, then continue to confirm. If none of the confirmed evaluation values ​​PDq exceed 20%, then directly generate a tensile test compliance signal and display it directly through the display unit.

7. The cloud computing-based intelligent assessment system for tensile fatigue of down jacket fabrics according to claim 1, characterized in that, The main analysis center handles segments without outliers in detail as follows: In subsequent cyclic tests, after the overall stretch length exceeded 30%, the tension was released, allowing the down jacket fabric to recover its elasticity. The recovery length between adjacent feature points was confirmed, as well as the original length associated with adjacent feature points before the first stretch test. The formula was: (Recovered length - Original length) ÷ Original length = BZ k Confirm the restoration rate BZ between corresponding adjacent feature points k Where k represents the connection segment between different feature points, and BZ represents several sets of recovery rates associated with different connection segments in the corresponding test process. k Select the maximum value BZ k max, if BZ k If the maximum deviation exceeds 20%, a tensile test failure signal is generated and displayed through the display unit. Otherwise, the tensile test continues, and the presence of BZ (Boundary Error) is assessed in subsequent tensile test processes. k Processes with a maximum percentage greater than 20%: If present, a tensile test failure signal will be generated directly.

8. The cloud computing-based intelligent assessment system for tensile fatigue of down jacket fabrics according to claim 7, characterized in that, If it does not exist, continue checking until the cyclical testing process reaches 30 sets, then stop. If none of them exist, BZ will be found. k For processes with a maximum completion rate greater than 20%, a tensile test compliance signal is directly generated and displayed directly through the display unit.

Images