Sneaker flaw detection system based on multi-source sensing data analysis

Abstract

The invention discloses a sneaker flaw detection system based on multi-source sensing data analysis, relates to the technical field of sneaker flaw detection, and aims to solve the technical problem that in the prior art, traditional detection only speculates a process through a wear form, but cannot quantify the influence of the defect on functional life. The system specifically comprises a flaw detection platform, and the flaw detection platform is in communication connection with a sole traceability wear detection unit, a suture line breakage detection unit and a crack path iteration prediction unit. The sole traceability wear detection unit is used for simulating a use scene of a sports shoe and detecting sole wear in the simulation process, after the sole traceability wear detection is completed, the suture line breakage detection unit is used for detecting suture line breakage of the sports shoe, and the crack path iteration prediction unit is used for iteratively predicting a crack path of a suture line.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sports shoes flaw detection, in particular to a sports shoes flaw detection system based on multi-source sensing data analysis. BACKGROUND

[0002] In the manufacturing process of sports shoes, the stitching line and the sole as the key load-bearing components, the quality defects directly affect the service life of the product. In the prior art, the traditional detection is only to speculate the process problem (such as the deviation of vulcanization temperature) through the wear pattern, but it cannot quantify the influence of the defect on the functional life (such as the rebound performance attenuation rate).

[0003] Therefore, the traditional detection process has surface problems: single wear analysis cannot associate the material dynamic performance attenuation, it is difficult to predict the actual service life of the defective sole, and also has deep problems: the cause-and-effect chain of process defects-microstructure changes-macro performance failure, and lacks quantitative model support for preventive improvement.

[0004] Therefore, the stitching line break detection of sports shoes is proposed. The traditional detection method only realizes the static identification of single defect, too much depends on the risk level of artificial experience judgment, and the conventional detection equipment cannot synchronously obtain acoustic, thermal and deformation data. Therefore, a solution is proposed. SUMMARY

[0005] The purpose of the present application is to solve the above-mentioned problems, and a sports shoes flaw detection system based on multi-source sensing data analysis is proposed.

[0006] The purpose of the present application can be realized by the following technical scheme: a sports shoes flaw detection system based on multi-source sensing data analysis, comprising a flaw detection platform, the flaw detection platform being communicatively connected with a sole traceable wear detection unit, a stitching line break detection unit and a crack path iterative prediction unit; The sole traceable wear detection unit simulates the use scene of sports shoes, and detects the sole wear during the simulation process; After completing the sole traceable wear detection, the stitching line break detection unit detects the stitching line break of the sports shoes; The crack path iterative prediction unit iteratively predicts the crack path of the stitching line.

[0007] As a preferred embodiment of the present application, the process of the sole traceable wear detection unit is as follows: According to the simulated use scene of the sole, the defect area of the sole is determined, the bearing pressure of the sole is collected, and the pressure distribution matrix is obtained by substituting the bearing pressure value according to the network distribution; Record the friction point position of the sole defect area in the simulation scene running time, determine the continuous frequency of the friction point position, and divide it into continuous friction point position and intermittent friction point position; In the process of positioning and collecting the friction point position, the temperature rise distribution of the sole defect area is captured, the friction heat production of the corresponding friction point position is obtained according to the temperature rise distribution, and the average heat production in the simulation use process is obtained by averaging. According to the pressure distribution matrix, the corresponding pressure value of the sole defect area is obtained, the pressure difference of different sole defect areas is obtained according to the pressure value deviation, and the pressure bearing deviation in the simulation use process is obtained by averaging.

[0008] As a preferred embodiment of the present application, if the pressure bearing deviation exceeds the pressure deviation threshold value, and the average heat production exceeds the average heat threshold value, it is inferred that the current sole has material shear fatigue defect, a shear fatigue defect signal is generated and sent to the defect detection platform; if the pressure bearing deviation does not exceed the pressure deviation threshold value, or the average heat production does not exceed the average heat threshold value, it is inferred that the current sole has process defect, that is, the process is traced back to the detection and targeted maintenance is carried out according to the trace result.

[0009] As a preferred embodiment of the present application, according to the continuous monitoring of the sole defect area, the deformation of the sole is counted and the deformation curve is constructed, and the dynamic resilience rate is calculated by the ratio of the rebound height, the compression depth and the maximum deformation in the deformation curve; The load-deformation phase diagram is constructed, the hysteresis loop area is calculated by integrating the closed path of the load-deformation phase diagram, the difference ratio of the reference dynamic resilience rate and the current dynamic resilience rate is multiplied by the ratio of the reference hysteresis loop area and the current hysteresis loop area; The difference ratio of the current dynamic resilience rate and the reference value is multiplied by the ratio of the current hysteresis loop area and the reference value to obtain the wear-resilience coupling coefficient; when the wear-resilience coupling coefficient exceeds the set coefficient threshold value, it is inferred that the defect causes the resilience performance to accelerate attenuation, otherwise, the wear resilience is continuously monitored.

[0010] As a preferred embodiment of the present application, the process of suture thread breakage detection unit is as follows: In the simulation use scene, the position of the suture thread is determined, which can be collected by image recognition; After determination, the acoustic signal and vibration signal are obtained according to the position; The acoustic vibration coherence energy of the suture thread data is calculated; The cross power spectral density function and the self power spectral density function of the acoustic signal and the vibration signal are calculated; The coherence coefficient is calculated according to the ratio of the cross power spectrum and the self power spectrum; The integral value of the coherence coefficient in the preset frequency band is extracted; The cross power spectral density function formula is embodied as: The self power spectral density function formula is embodied as: And ; Gxx(f) is the self power spectral density; Gxy(f) is the cross power spectral density; The coherence coefficient formula is embodied as: ; The coherence coefficient is integrated in the preset frequency band [f1, f2], ; In the preset frequency, the area integral value A exceeds the set integral threshold value, so that the characteristic frequency caused by the defect is more significant, and a high-impact signal of the defect is generated.

[0011] As a preferred embodiment of the present application, the defect risk signal and the defect high-impact signal are sent to the flaw detection platform. After receiving the defect risk signal, the flaw detection platform repairs the stitching line without interrupting the use process of the simulated use scene, and is used for further detection of the performance of the sports shoes in this scene. After receiving the defect high-impact signal, the use process of the simulated use scene is interrupted and repaired, and the wiring track of the current stitching line connection is detected.

[0012] As a preferred embodiment of the present application, the process of the crack path iteration prediction unit is as follows: When the stitching line breakage detection is abnormal, the crack of the stitching line is collected, and the equivalent length of the crack is processed, and the equivalent crack length is set as a label a; The stress intensity factor is calculated, and the calculation formula is: ; P is the stress intensity factor, and Y is the geometric factor; is the external stress borne by the component; If P does not exceed the factor threshold value, it is inferred that the crack is stable, otherwise, if P exceeds the factor threshold value, it is inferred that the crack is unstable.

[0013] As a preferred embodiment of the present application, when the crack is unstable, the path iteration is performed along the maximum principal stress direction: Through stress field analysis, the maximum principal stress direction of the crack tip is found, a micro-increment is expanded each time, and a is recorded 增 ; The equivalent crack length after the increase is used to recalculate P, and threshold comparison is performed again. If P still exceeds the factor threshold value, the expansion along the maximum principal stress direction is continued, and the length increment is updated. The calculation is repeated until the termination condition is met. The condition is that the equivalent crack length exceeds the safety red line value, or the stress intensity factor is lower than the factor threshold value; According to the crack propagation path obtained by the extension azimuth and length, the complete path of crack propagation is sent to the flaw detection platform; the flaw detection platform performs suture repair according to the path after receiving, and adjusts the processing technology of the running shoes, strengthens the toughness of the suture used at the corresponding path position, or improves the process to reduce the stress of the suture.

[0014] Compared with the prior art, the beneficial effects of the present application are: 1、In the present application, the traditional detection only speculates the process problem (such as sulfurization temperature deviation) through the wear pattern, the present application is beneficial to quantifying the influence of the defect on the functional life (such as the rebound performance attenuation rate), and the material dynamic performance attenuation can be associated in single wear analysis, the actual service life of the defective sole can be accurately predicted, and the cause-effect chain of process defect-microstructure change-macro performance failure is ensured, and the quantitative model is supported for preventive improvement.

[0015] 2、In the present application, in addition to realizing the static identification of single defect, defect prediction and evolution analysis are realized through multi-source data fusion, a quantitative evaluation system based on a mechanical model is established, multi-dimensional information space correlation analysis is realized through a sensor array, potential fracture risk of a high stress concentration area of the suture is identified in advance, a crack propagation path is accurately predicted, and data support is provided for process improvement.

[0016] 3、In the present application, the crack path of the suture is iteratively predicted, so that the crack growth trend and the critical fracture point cannot be predicted in the flaw detection process, the accuracy of the flaw detection is effectively and accurately improved, the prediction can be performed in time, the crack risk protection is effectively performed, the crack generation rate is reduced, and the scrap rate of the sports shoes is reduced. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to facilitate those skilled in the art to understand, the present application will be further described below with reference to the accompanying drawings.

[0018] Fig. 1 The system principle block diagram of the present application is shown in the figure; Fig. 2 The method flow chart of the present application is shown in the figure. DETAILED DESCRIPTION

[0019] In order to make the personnel in the technical field better understand the present application scheme, the technical scheme in the embodiment of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiment of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor belong to the scope of protection of the present application.

[0020] Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are they necessarily mutually exclusive of one another. As will be apparent to those of ordinary skill in the art, embodiments described herein can be combinable with other embodiments.

[0021] Referring to Figs. 1-2 As shown, the sneaker flaw detection system based on multi-source sensor data analysis comprises a flaw detection platform, the flaw detection platform is in communication connection with a sole trace wear detection unit, a suture thread breakage detection unit and a crack path iterative prediction unit; The flaw detection platform generates a sole trace wear detection signal and sends it to the sole trace wear detection unit; The sole trace wear detection unit simulates the use scene of the sneaker, and detects the sole wear during the simulation process. In the prior art, the traditional detection is only to infer the process problem (such as sulfurization temperature deviation) through wear morphology, but it cannot quantify the impact of the defect on the functional service life (such as the rebound performance attenuation rate); Therefore, the traditional detection process has surface problems: single wear analysis cannot associate material dynamic performance attenuation, and it is difficult to predict the actual service life of the defective sole; there are also deep-seated problems: the causal chain of process defects-microstructure changes-macro performance failure is broken, and there is a lack of quantitative model to support preventive improvement.

[0022] According to the simulated use scene of the sole, the defective area of the sole is determined, the bearing pressure of the sole is collected through the pressure sensor, and a pressure sensor network is constructed, and the pressure distribution matrix is obtained according to the network distribution and the bearing pressure value; The friction point position of the defective area of the sole during the running of the simulation scene is recorded by the acoustic emission sensor. After the friction point position is located, the continuous frequency of the friction point position in the defective area of the sole is determined, and the continuous frequency is used as the friction point position type division standard, that is, if the continuous frequency exceeds the set frequency threshold, it is marked as a continuous friction point position, otherwise, it is marked as a discontinuous friction point position; During the positioning and collection of the friction point position, the temperature rise distribution of the defective area of the sole is captured, the friction heat generated by the corresponding friction point position is obtained according to the temperature rise distribution, and the average value is obtained to obtain the heat generation average value during the simulation use process; According to the pressure distribution matrix, the corresponding pressure value of the defective area of the sole is obtained, the pressure difference of different defective areas of the sole is obtained according to the pressure value deviation, and the average value is obtained to obtain the pressure bearing deviation during the simulation use process; If the pressure bearing deviation exceeds the pressure deviation threshold value, and the heat generation average value exceeds the heat average threshold value, it is inferred that the current shoe sole has a material shear fatigue defect, a shear fatigue defect signal is generated and sent to the defect detection platform; If the pressure bearing deviation does not exceed the pressure deviation threshold value, or the heat generation average value does not exceed the heat average threshold value, it is inferred that the current shoe sole has a process defect, i.e. the process is traced and maintained according to the trace result; According to the continuous monitoring of the shoe sole defect area, the deformation of the shoe sole is counted and a deformation curve is constructed, and the dynamic resilience rate is calculated by the ratio of the rebound height, the compression depth and the maximum deformation in the deformation curve; A load-deformation phase diagram is constructed, the hysteresis loop area is calculated by integrating the closed path of the load-deformation phase diagram, the difference ratio of the reference dynamic resilience rate and the current dynamic resilience rate is multiplied by the ratio of the reference hysteresis loop area and the current hysteresis loop area; The dynamic resilience rate refers to the quantitative index of the deformation recovery ability of the material under dynamic load, which can be realized by normalizing the ratio of the rebound height and the compression depth in the deformation curve combined with the maximum deformation, and this parameter is used to represent the elastic retention ability of the material after repeated stress; The hysteresis loop area refers to the energy loss in the load-deformation cycle, which can be realized by numerical integration of the closed curve formed by load and deformation, and this parameter reflects the energy dissipation degree of the material; The reference dynamic resilience rate and the reference hysteresis loop area refer to the reference values established by standard sample testing, which can use the average value of historical test data of the same batch of qualified products as the reference value to quantify the deviation of the current product performance; Specifically, when detecting the shoe sole defect, first, the dynamic deformation-load data is synchronously collected by the pressure sensor and the laser displacement meter to generate a load-deformation phase diagram, the rebound height is extracted from the deformation curve in the unloading stage, the compression depth is the maximum deformation in the loading stage, and the product of the ratio of the two and the maximum deformation constitutes the dynamic resilience rate. The hysteresis loop area is obtained by numerical integration of the closed path in the phase diagram; The difference ratio of the current dynamic resilience rate and the reference value is multiplied by the ratio of the current hysteresis loop area and the reference value to obtain the wear-resilience coupling coefficient; This coefficient comprehensively reflects the coupling effect of material elasticity attenuation and energy loss, and triggers the remaining life calculation when it exceeds the set threshold value; When the wear-resilience coupling coefficient exceeds the set coefficient threshold value, it is inferred that the defect causes the resilience performance to accelerate attenuation, otherwise, the wear resilience is continuously monitored; Compared with the prior art, the traditional method only evaluates the sole performance by a single static parameter, such as only measuring the wear depth or the rebound height, while the method is coupled with the dynamic rebound rate and the hysteresis area to simultaneously capture the dual effects of the decrease in the material elastic recovery capability and the accumulation of internal structure damage, and no dynamic comparison mechanism of the baseline value and the current value is established in the prior art, so that the performance degradation degree cannot be quantified, while the method is operated by the product of the difference ratio and the area ratio to accurately quantify the material state change.

[0023] After the sole traceability wear detection is completed, a stitching thread breakage detection signal is generated and sent to a stitching thread breakage detection unit; After the stitching thread breakage detection unit receives the stitching thread breakage detection signal, the sports shoes are subjected to stitching thread breakage detection, and the traditional detection method only realizes static identification of a single defect, while the scheme realizes defect prediction and evolution analysis through multi-source data fusion; the traditional method relies on artificial experience to judge the risk level, and the scheme establishes a quantitative evaluation system based on a mechanical model, and conventional detection equipment cannot simultaneously acquire acoustic, thermal and deformation data, and the scheme realizes spatio-temporal correlation analysis of multi-dimensional information through a sensor array; That is, through the above technical scheme, the application can identify the potential fracture risk of the high stress concentration area of the stitching thread in advance, accurately predict the crack propagation path, and provide data support for process improvement.

[0024] In the simulation use scenario, the position of the stitching thread is determined, and the determination method can be collected through image recognition; After the position is determined, acoustic signals and vibration signals are acquired according to the position; The acoustic vibration coherence energy of the stitching thread data is calculated; The cross-power spectral density function and the self-power spectral density function of the acoustic signal and the vibration signal are calculated; The coherence coefficient is calculated according to the ratio relationship of the cross-power spectrum and the self-power spectrum; The integral value of the coherence coefficient in the preset frequency band is extracted; The acoustic signal is collected by a sensor, the acoustic wave of the stitching thread vibration (such as friction, structure deformation noise), the vibration signal is collected by an acceleration sensor, the mechanical vibration of the stitching thread (such as deformation, stress release vibration), and the signals are synchronously collected (time alignment) to ensure that the acoustic and vibration signals are responses of the same physical process.

[0025] The cross-power spectral density function is a cross-correlation function of two signals in the frequency domain, which can be realized by performing frequency spectrum analysis on the acoustic signal and the vibration signal by using a fast Fourier transform algorithm, and is used to represent the energy coupling relationship of the acoustic and vibration signals at a specific frequency; The formula is as follows: The self-power spectral density function refers to the energy distribution of a single signal in the frequency domain, which can be specifically realized by performing Fourier transform on the autocorrelation function of the acoustic signal or the vibration signal, and is used to reflect the frequency spectrum characteristics of the signals themselves; The formula is embodied as: Or The formula is explained as: f is the frequency, usually in hertz (Hz), which is a coordinate axis describing the signal in the frequency domain, representing the "position" of different frequency components in the signal, for example, the vibration component of 200 Hz in the acoustic signal is embodied at the frequency point f = 200; FFT[·] is the operator symbol of fast Fourier transform, which is a mathematical algorithm, and its function is to convert the time-domain signal (signal with time as the variable) to the frequency domain (signal represented by frequency as the variable), which is convenient for analyzing the energy, phase and other characteristics of different frequency components in the signal, and is the core operation of signal processing from time domain to frequency domain analysis; x(t), y(t): t represents time, which is the independent variable of the time-domain signal, describing the law of signal change over time; x(t) refers to the acoustic signal (such as the function of the sound signal collected in the suture detection over time); y(t) refers to the vibration signal (function of suture vibration displacement, acceleration, etc. over time), which is a time-domain signal, recording the fluctuation of physical quantities in the time dimension; x*(t) y*(t) is a complex conjugate operation; If x(t) is a complex time-domain signal (in actual engineering, for the convenience of mathematical processing, real signals are often represented in complex form, including real and imaginary parts), x*(t) is the signal after taking the negative of the imaginary part. In power spectral density calculation, the conjugate operation can accurately calculate the energy and correlation characteristics of the signal, ensuring that the calculated power spectrum is a real number and has physical meaning (representing energy distribution); Gxx(f) and Gyy(f) are self-power spectral densities, and Gxy(f) is mutual power spectral density.

[0026] The coherence coefficient refers to the linear correlation degree of two signals in the frequency domain, which can be specifically realized by calculating the ratio of the square of the mutual power spectrum modulus to the product of the two self-power spectra, and is used to judge the causal relationship between the acoustic and vibration signals at a specific frequency; The formula is embodied as: The value range of is between 0 and 1, if is approximately equal to 1, the acoustic signal and the vibration signal are strongly correlated at this frequency, and the defect risk signal is generated; On the contrary, if ​About equal to 1, the acoustic signal, vibration signal is irrelevant at this frequency, set as noise interference.

[0027] The integral value of the coherence coefficient in the preset frequency band refers to the area integral of the coherence coefficient in a specific frequency range, which can be realized by setting a frequency band range matching the inherent frequency of the suture material, and is used for focusing on the characteristic frequency response caused by the abnormal structure of the material; Specifically: integrating the coherence coefficient in the preset frequency band [f1, f2], ; In the preset frequency, the area integral value A exceeds the set integral threshold value, then it is inferred that the characteristic frequency caused by the defect is more significant, and a high-impact signal of the defect is generated; The defect risk signal and the high-impact signal of the defect are sent to the flaw detection platform, and after receiving the defect risk signal, the suture is repaired without interrupting the use process of the simulated use scene, which is used for further detection of the performance of the sports shoes in this scene, and after receiving the high-impact signal of the defect, the use process of the simulated use scene is interrupted and repaired, and the wiring track of the connection of the current suture is detected; Meanwhile, a crack path iterative prediction signal is generated and sent to a crack path iterative prediction unit; The crack path iterative prediction unit iteratively predicts the crack path of the suture to solve the problem that the crack growth trend and the critical breaking point cannot be predicted in the flaw detection process, effectively and accurately improves the accuracy of the flaw detection, timely prediction, effective crack risk protection, reduces the crack generation rate, and reduces the scrap rate of sports shoes.

[0028] When the suture detection anomaly occurs, the crack of the suture is collected, and the equivalent crack length is processed, and the equivalent crack length is set as a label a; The role of the equivalent crack length is to "equivalent conversion" these irregular cracks into a straight line crack, so that different forms of cracks can be quantified by the same parameter; The specific processing method is: based on the stress field distribution of the crack tip; no matter how the crack shape is, the stress concentration effect (such as stress size, distribution range) of its tip is the core feature, through numerical calculation (such as finite element analysis), find a "virtual straight line crack", so that the stress field distribution of its tip is consistent with the actual crack, the length of this virtual crack is the equivalent crack length; Stress intensity factor calculation, stress intensity factor is the core parameter of fracture mechanics, which is used to quantify the stress concentration intensity of the crack tip, the larger the value, the easier the crack to expand; The calculation formula is: ; P is a stress intensity factor, Y is a geometric factor (related to crack location, component shape, such as edge crack Y is about 1.12, central crack is about 1.0); is the external stress suffered by the component, such as the tensile stress suffered by the suture, unit: MPa; If P does not exceed the factor threshold, it is concluded that the crack is stable, otherwise, if P exceeds the factor threshold, it is concluded that the crack is unstable; When the crack is unstable, path iteration is performed: Along the maximum principal stress direction (the material is most easily torn in this direction), the maximum principal stress direction of the crack tip is found through stress field analysis, and a small increment is expanded each time, and a is recorded 增 ; Recalculate P with the increased equivalent crack length, and compare the threshold value again, if P still exceeds the factor threshold, continue to expand along the maximum principal stress direction, and update the length increment, repeat the calculation until the termination condition is met, the condition is that the equivalent crack length exceeds the safety red line value, or the stress intensity factor is lower than the factor threshold; According to the expansion direction and length, the complete path of crack expansion is obtained, and is sent to the defect detection platform; The defect detection platform receives and repairs the suture according to the path, and adjusts the processing technology of the running shoes, strengthens the toughness of the suture used at the corresponding path position, or improves the process to reduce the stress of the suture.

[0029] In use, the sole traceability wear detection unit simulates the use scene of the sports shoes, and detects the sole wear during the simulation process, after completing the sole traceability wear detection, the suture broken wire detection unit detects the suture broken wire of the sports shoes, and the crack path iteration prediction unit iterates and predicts the crack path of the suture.

[0030] The threshold or the preset value, the preset range and the like are set for result comparison and analysis, so as to determine whether it is good or bad, and the size of the threshold is set according to the large model analysis of sample data and artificial experience, and is also stored by recording, and can be appropriately adjusted through seasonal or common influence conditions; And the weight proportion coefficient, the influence factor and the like are set according to the influence size of each parameter on the result, to allocate specific numerical values to finally reflect the influence condition on the result, which is also set by recording storage through large model analysis of sample data and artificial experience, and can be appropriately adjusted through seasonal or common influence conditions.

[0031] The preferred embodiments of the application disclosed above are only to facilitate the elucidation of the application. The preferred embodiments do not describe all the details of the application and limit the application to the specific embodiments. Obviously, many modifications and variations can be made in light of the teachings above. The description is chosen and described in order to provide the best illustration of the application and its practical application to those skilled in the art and to enable those skilled in the art to best utilize the application. The application is limited only by the claims and their full scope and equivalents.

Claims

1. A sports shoe defect detection system based on multi-source sensor data analysis, comprising a defect detection platform, characterized in that, The defect detection platform has communication connections for a shoe sole wear detection unit, a stitching thread breakage detection unit, and a crack path iterative prediction unit. The sole wear detection unit simulates usage scenarios for athletic shoes and detects sole wear during the simulation process. After completing the sole wear test, the stitching breakage detection unit performs stitching breakage detection on the athletic shoes. The crack path iterative prediction unit performs iterative prediction of the crack path of the suture line.

2. The sports shoe defect detection system based on multi-source sensor data analysis according to claim 1, characterized in that, The process of the sole wear detection unit is as follows: Based on the simulated usage scenario of the shoe sole, the defect area of ​​the shoe sole is determined, the load-bearing pressure of the shoe sole is collected, and the load-bearing pressure value is substituted into the network distribution to obtain the pressure distribution matrix. Record the friction points in the defective area of ​​the shoe sole during the simulation scenario, determine the duration of the friction points, and divide them into continuous friction points and intermittent friction points; During the location and collection of friction points, the temperature rise distribution of the defective area of ​​the sole is captured. The friction heat generated at the corresponding friction point is obtained based on the temperature rise distribution, and the average value is calculated to obtain the average heat generation during the simulated use process. The pressure values ​​corresponding to the defective areas of the sole are obtained from the pressure distribution matrix. The pressure difference between different defective areas of the sole is obtained from the pressure value deviation. The average value is then used to obtain the pressure bearing deviation during the simulated use process.

3. The sports shoe defect detection system based on multi-source sensor data analysis according to claim 2, characterized in that, If the pressure bearing deviation exceeds the pressure deviation threshold and the average heat generation exceeds the average heat generation threshold, it is inferred that there is a material shear fatigue defect in the current sole, and a shear fatigue defect signal is generated and sent to the defect detection platform; if the pressure bearing deviation does not exceed the pressure deviation threshold or the average heat generation does not exceed the average heat generation threshold, it is inferred that there is a process defect in the current sole, that is, the process is traced and detected, and targeted maintenance is carried out based on the traceability results.

4. The sports shoe defect detection system based on multi-source sensor data analysis according to claim 3, characterized in that, Based on continuous monitoring of defect areas in the sole, the deformation of the sole is statistically analyzed and a deformation curve is constructed. The dynamic rebound rate is calculated by the ratio of the rebound height, compression depth and maximum deformation in the deformation curve. Construct a load-deformation phase diagram, calculate the hysteresis loop area by integrating the closed path of the load-deformation phase diagram, and multiply the ratio of the difference between the reference dynamic rebound rate and the current dynamic rebound rate by the ratio of the reference hysteresis loop area to the current hysteresis loop area. The wear-rebound coupling coefficient is obtained by multiplying the ratio of the difference between the current dynamic rebound rate and the reference value by the ratio of the current hysteresis loop area to the reference value. When the wear-rebound coupling coefficient exceeds the set coefficient threshold, it is inferred that the defect causes the rebound performance to decay faster. Otherwise, the wear rebound is continuously monitored.

5. The sports shoe defect detection system based on multi-source sensor data analysis according to claim 4, characterized in that, The process of the suture breakage detection unit is as follows: In simulated usage scenarios, the location of the suture line is determined, specifically through image recognition. Once determined, acoustic and vibration signals are acquired based on the location. Calculate the acoustic-vibrational coherent energy acquisition from suture data; Calculate the cross-power spectral density function and the self-power spectral density function of the acoustic signal and the vibration signal; The coherence coefficient is calculated based on the ratio of the cross power spectrum to the self power spectrum. Extract the integral value of the coherence coefficient within the preset frequency band; The cross-power spectral density function formula is expressed as follows: ; The formula for the self-power spectral density function is as follows: and ; Gxx(f) and Gyy(f) are the self-power spectral densities; Gxy(f) is the cross-power spectral density; The formula for the coherence coefficient is as follows: ; Integrating the coherence coefficient over the preset frequency band [f1, f2], ; Within a preset frequency range, if the area integral value A exceeds the set integration threshold, it is inferred that the characteristic frequency caused by the defect is more significant, resulting in a high-impact defect signal.

6. The sports shoe defect detection system based on multi-source sensor data analysis according to claim 5, characterized in that, The defect risk signal and the high impact signal of the defect are sent to the defect detection platform. After receiving the defect risk signal, the defect detection platform repairs the suture without stopping the use of the simulated use scenario. Upon receiving a high-impact defect signal, the simulation of the usage scenario is stopped and repairs are performed, while the current suture connection is checked for its routing trajectory.

7. The sports shoe defect detection system based on multi-source sensor data analysis according to claim 6, characterized in that, The process of the crack path iterative prediction unit is as follows: When the suture breakage detection is abnormal, the crack of the suture is collected, and the equivalent length of the crack is processed, and the equivalent crack length is labeled a; The stress intensity factor is calculated using the following formula: ; P is the stress intensity factor, and Y is the geometric factor; The external stress applied to the component; If P does not exceed the factor threshold, the fracture is stable; otherwise, if P exceeds the factor threshold, the fracture is unstable.

8. The sports shoe defect detection system based on multi-source sensor data analysis according to claim 7, characterized in that, When the crack is unstable, path iteration is performed along the direction of maximum principal stress: By analyzing the stress field, the direction of the maximum principal stress at the crack tip was found. Each time the crack expanded by a small increment, the value of a was recorded. 增 ; Recalculate P using the increased equivalent crack length, then compare the thresholds. If P still exceeds the factor threshold, continue to extend along the direction of the maximum principal stress and update the length increment. Repeat the calculation until the termination condition is met: the equivalent crack length exceeds the safety red line value or the stress intensity factor is lower than the factor threshold. The complete path of crack propagation is obtained based on the direction and length of the crack and sent to the defect detection platform. After receiving the crack, the defect detection platform repairs the sutures according to the path and adjusts the processing technology of the running shoe to strengthen the toughness of the sutures used at the corresponding path location or improve the process to reduce the stress on the sutures.

Images