A method for analyzing hidden damage of a conveyor belt by using magnetic characteristics

Abstract

The present invention relates to the technical field of material physical property detection for hidden damage of conveyor belts, and discloses a method for analyzing hidden damage of conveyor belts by using magnetic characteristics, including: obtaining hysteresis loop data in real time, identifying the slope inflection points in the magnetization and demagnetization stages as reference marks, correcting the morphological distortion caused by speed fluctuations through a phase synchronization mechanism based on the dynamic mapping relationship between the time interval and the amplitude ratio, and realizing damage determination by combining energy integral correction and gradient direction analysis. The present invention uses the inherent phase attribute of the hysteresis loop to decouple the speed interference and the damage signal, constructs a full-element analysis logic for damage characteristics through multi-dimensional data collaboration, and significantly improves the detection confidence and positioning accuracy under complex working conditions; at the same time, based on single-sensor in-depth analysis and dynamic reference self-adaptive calibration, all-weather stable monitoring is realized with zero hardware modification cost, providing a highly robust and low-cost hidden damage detection solution for heavy industrial scenarios such as mines and ports.

Description

Technical Field

[0001] The present invention relates to the technical field of detecting the physical properties of materials in the aspect of hidden damage of conveyor belts, and in particular to a method for analyzing hidden damage of conveyor belts by using magnetic characteristics. Background Art

[0002] In the quality control and safety detection of conveyor belts, identifying potential defects by measuring the physical properties of materials is an important technical means. In the prior art, for the magnetic characteristic analysis method of conveyor belt damage, it is mainly to measure and analyze the magnetic response presented by the conveyor belt under the action of a magnetic field, that is, the hysteresis loop. The hysteresis loop can reflect the magnetization characteristics of materials, and physical defects inside the conveyor belt material, such as cracks, fractures or non-uniformities, etc., will directly affect its magnetic behavior and show specific changes or distortions on the hysteresis loop. By extracting and analyzing the characteristics of these hysteresis loops, the physical state of the conveyor belt can be evaluated and whether there is damage can be judged.

[0003] However, in practical applications, especially in the on-line detection scenario of conveyor belts, a key technical challenge lies in how to accurately measure and analyze its inherent magnetic physical properties without being interfered by external factors. Among them, the dynamic fluctuation of the conveyor belt running speed is a significant interference factor. Due to reasons such as load changes and motor speed regulation, the speed of the conveyor belt is not constant, and this speed change will directly affect the magnetization process and demagnetization process of the conveyor belt material, resulting in the distortion of the measured hysteresis loop shape. This change in the hysteresis loop shape caused by speed fluctuations has similarities with the distortion caused by the physical defects of the material itself in some aspects, which brings serious troubles to the damage analysis based on magnetic physical properties and is prone to misjudging the true physical state of the conveyor belt. In order to reduce the influence of speed fluctuations on the measurement of magnetic physical properties, the commonly used method in the prior art is:

[0004] 1. Sacrifice the measurement sensitivity. By reducing the sensitivity of the magnetic detection system, the influence degree of speed change on the shape of the hysteresis loop can be reduced, thereby reducing the risk of misjudgment. However, the direct consequence of this approach is that for the subtle physical damage in the conveyor belt, the resulting change in magnetic characteristics may be small and difficult to be effectively captured under the measurement conditions of low sensitivity, leading to the failure to detect potential minor damages in a timely manner, which is obviously not conducive to the comprehensive assessment and safety guarantee of the physical state of the conveyor belt. 2. There are also some existing technical solutions that attempt to introduce additional speed measurement devices, such as speed sensors, to obtain the real-time speed information of the conveyor belt and establish a mathematical model based on these speed data to compensate or correct the measured hysteresis loop. These methods attempt to separate the influence of speed change from the measured magnetic physical data, so as to more accurately reflect the true magnetic physical properties of the conveyor belt material. However, these solutions also have some inherent problems, such as increasing the complexity and cost of the test system: additional sensors and complex signal processing units will undoubtedly increase the hardware cost and system integration difficulty, and establishing an accurate speed compensation model often requires a large amount of experimental data and complex algorithms. In the case of rapid speed change of the conveyor belt, the accuracy and real-time performance of the model are difficult to be guaranteed, and it may not be able to truly reflect the instantaneous magnetic physical properties of the material. Moreover, the additional sensors themselves may also have measurement errors and may be affected by environmental factors, thus having an adverse impact on the final analysis results of the physical properties. Summary of the Invention

[0005] The technical problem solved by the present invention is to provide a method for analyzing latent damage of a conveyor belt by using magnetic characteristics to solve the problems of the distortion of the hysteresis loop shape caused by the dynamic fluctuation of the running speed of the conveyor belt and the high implementation cost and poor real-time performance of the traditional multi-sensor scheme in the above-mentioned prior art.

[0006] To solve the above technical problems, the technical solution adopted by the present invention is as follows: A method for analyzing latent damage of a conveyor belt by using magnetic characteristics includes the following steps:

[0007] Step S1, obtaining the hysteresis loop data during the operation of the conveyor belt in real time, where the hysteresis loop data includes the corresponding relationship between the magnetic field strength and the time series in the magnetization stage and the demagnetization stage;

[0008] Step S2, identifying the first slope inflection point as the magnetization reference mark in the magnetization stage, and the first slope inflection point is the critical point where the slope first changes suddenly in the rising edge of the magnetic field strength changing with time;

[0009] Step S3, identifying the second slope inflection point as the demagnetization reference mark in the demagnetization stage, and the second slope inflection point is the critical point where the slope first changes suddenly in the falling edge of the magnetic field strength changing with time;

[0010] Step S4. Calculate the magnetic field strength amplitude ratio between the magnetization reference mark and the demagnetization reference mark according to the time interval therebetween , where is the magnetic field strength amplitude corresponding to the magnetization reference mark, is the magnetic field strength amplitude corresponding to the demagnetization reference mark, and the amplitude ratio satisfies the following relationship:

[0011] ,

[0012] where is a preset phase synchronization constant, is the reference time interval of the conveyor belt at the standard running speed;

[0013] Step S5. Based on the mapping relationship between the amplitude ratio and the preset phase synchronization reference line, dynamically adjust the time axis ratio and amplitude axis ratio of the hysteresis loop so that the slope inflection points of the adjusted magnetization stage and demagnetization stage are aligned with the phase synchronization reference line;

[0014] Step S6. Output the corrected hysteresis loop data, and determine the hidden damage position of the conveyor belt according to the abnormal distortion characteristics of the magnetic field strength distribution in the corrected data.

[0015] As a further solution of the present invention, the method for identifying the first slope inflection point in step S2 includes: calculating the first derivative of the curve of the magnetic field strength varying with time in the magnetization stage, and defining the position where the derivative first drops below 20% of the peak value as the first slope inflection point.

[0016] As a further solution of the present invention, the dynamic adjustment in step S5 includes: when is less than the preset threshold, compress the time axis ratio of the hysteresis loop and amplify the amplitude axis ratio; when is greater than the preset threshold, stretch the time axis ratio and reduce the amplitude axis ratio until the time difference between the magnetization reference mark and the demagnetization reference mark reaches the preset synchronization constant.

[0017] As a further solution of the present invention, the method for setting the phase synchronization reference line includes: collecting the spatio-temporal distribution characteristics of the magnetization reference mark and the demagnetization reference mark of the hysteresis loop during operation at the standard speed in the non-damaged state of the conveyor belt, and calculating the synchronization constant through the following formula:

[0018] ,

[0019] Among them, is the magnetic field strength amplitude of the magnetization reference mark in the non-damaged state, is the magnetic field strength amplitude of the demagnetization reference mark in the non-damaged state, is the reference time interval between the two.

[0020] As a further solution of the present invention, the abnormal distortion feature determination method in step S6 includes: calculating the local curvature of the corrected hysteresis loop, marking the continuous region with a curvature exceeding the preset threshold as a potential damage region, and performing magnetic field strength gradient analysis on this region. If the gradient change rate exceeds 3 times the historical reference value, it is determined as a hidden damage.

[0021] As a further solution of the present invention, the method further includes a dynamic phase synchronization verification step: in the corrected hysteresis loop data, if the slope inflection points of the magnetization stage and the demagnetization stage are not completely aligned to the phase synchronization reference line, repeat steps S2 to S5 until the time difference and the amplitude ratio have a product error less than 5%.

[0022] As a further solution of the present invention, the calculation method of the amplitude ratio in step S4 includes: performing sliding window mean filtering on the magnetic field strength data of the magnetization reference mark and the demagnetization reference mark, and taking 80% of the filtered peak value as the and calculation reference values.

[0023] As a further solution of the present invention, the method further includes a damage verification step: when a hidden damage is detected, trigger the conveyor belt to decelerate to 50% of the standard speed, re-collect the hysteresis loop data and perform secondary correction. If the damage feature still exists after the secondary correction, generate a damage alarm signal.

[0024] As a further solution of the present invention, the method for obtaining the hysteresis loop data includes: symmetrically arranging a magnetic induction probe array on both sides of the conveyor belt, collecting magnetic field strength distribution data in real time, and eliminating environmental magnetic field interference through differential calculation.

[0025] ​Compared with the prior art, the beneficial effects of the present invention are as follows: Based on the spatio-temporal alignment characteristics of the slope inflection points in the magnetization and demagnetization stages of the hysteresis loop, the morphological distortion caused by speed fluctuations is converted into reversible correction parameters. Through endogenous phase difference perception and dynamic scaling of the time axis, physical separation of speed interference and damage signals is achieved. Since this mechanism does not rely on external sensors or complex compensation models, it fundamentally solves the misjudgment problem caused by speed-damage coupling in traditional methods in dynamic speed regulation scenarios. By integrating the time evolution characteristics (inflection point amplitude ratio), energy integral difference, and gradient direction distribution of the hysteresis loop, a multi-dimensional cross-validation logic for damage determination is constructed. For example, the energy integral difference compensates for the interference of load mutations on the magnetization energy, and the gradient direction maps the material failure modes (gradient attenuation of surface cracks and step mutations of internal delamination), achieving a significant improvement in the detection confidence under complex working conditions through multi-dimensional collaboration. It is possible to avoid relying on multi-sensor fusion in hardware. For example, spatio-temporal multi-dimensional data analysis (time axis stretching, amplitude axis scaling, energy integral calculation) using a single magnetic sensor is used to achieve all-element detection of hidden damage to the conveyor belt. This path can directly reuse the existing magnetic induction equipment on the production line, avoiding the cost of hardware transformation, providing a plug-and-play detection upgrade solution for heavy industry scenarios such as mines and ports. At the same time, a lightweight inflection point detection algorithm (such as first derivative threshold truncation) and a closed-loop verification process are adopted to achieve real-time processing and anti-electromagnetic interference capabilities under extreme working conditions such as sudden conveyor belt stops and load mutations, ensuring stable all-weather operation in strong noise environments such as mines and metallurgy. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description 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.

[0027] Figure 1 It is a flowchart of the method for analyzing hidden damage to the conveyor belt using magnetic characteristics of the present invention.

[0028] Figure 2 It is a flowchart of the dynamic phase synchronization correction of the present invention.

[0029] Figure 3 It is a flowchart of hidden damage determination based on local curvature and magnetic field gradient of the present invention.

[0030] Figure 4 It is a structural diagram of the functional modules of the hidden damage detection system for the conveyor belt of the present invention.

[0031] Figure 5 It is a schematic diagram of the structure of the multi-dimensional damage analysis module of the present invention. Detailed implementation manners

[0032] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Apparently, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0033] Please refer to Figure 1 , a method for analyzing hidden damage of a conveyor belt by using magnetic characteristics, including the following steps:

[0034] Step S1: Real-time obtain the hysteresis loop data during the operation of the conveyor belt. The hysteresis loop data includes the corresponding relationship between the magnetic field strength and the time series in the magnetization stage and the demagnetization stage;

[0035] Step S2: Identify the first slope inflection point as the magnetization reference mark in the magnetization stage. The first slope inflection point is the critical point where the slope first changes suddenly in the rising edge of the magnetic field strength changing with time;

[0036] Step S3: Identify the second slope inflection point as the demagnetization reference mark in the demagnetization stage. The second slope inflection point is the critical point where the slope first changes suddenly in the falling edge of the magnetic field strength changing with time;

[0037] Step S4: According to the time interval between the magnetization reference mark and the demagnetization reference mark , calculate the magnetic field strength amplitude ratio of the two, where is the magnetic field strength amplitude corresponding to the magnetization reference mark,

[0038] is the magnetic field strength amplitude corresponding to the demagnetization reference mark, and the amplitude ratio satisfies the following relationship:

[0039] where, is a preset phase synchronization constant, is the reference time interval of the conveyor belt at the standard running speed;

[0040] Step S5: Based on the mapping relationship between the amplitude ratio and the preset phase synchronization reference line, dynamically adjust the time axis ratio and amplitude axis ratio of the hysteresis loop so that the slope inflection points of the adjusted magnetization stage and demagnetization stage are aligned to the phase synchronization reference line;

[0041] Step S6: Output the corrected hysteresis loop data, and determine the hidden damage position of the conveyor belt according to the abnormal distortion characteristics of the magnetic field strength distribution in the corrected data.

[0042] As a further solution of the present invention, the method for identifying the first slope inflection point in step S2 includes: calculating the first derivative of the curve of the magnetic field strength varying with time in the magnetization stage, and defining the position where the derivative first drops below 20% of the peak value as the first slope inflection point.

[0043] As a further solution of the present invention, the dynamic adjustment in step S5 includes: when is less than a preset threshold, compressing the time axis ratio of the hysteresis loop and enlarging the amplitude axis ratio; when is greater than the preset threshold, stretching the time axis ratio and reducing the amplitude axis ratio until the time difference between the magnetization reference mark and the demagnetization reference mark and the amplitude ratio

[0044] reach a preset synchronization constant. :

[0045] ,

[0046] where is the magnetic field strength amplitude of the magnetization reference mark in the non-damaged state, is the magnetic field strength amplitude of the demagnetization reference mark in the non-damaged state, is the reference time interval between the two.

[0047] As a further solution of the present invention, the method for determining the abnormal distortion feature in step S6 includes: calculating the local curvature of the corrected hysteresis loop, marking the continuous region where the curvature exceeds the preset threshold as the potential damage region, and performing magnetic field strength gradient analysis on this region. If the gradient change rate exceeds 3 times the historical reference value, it is determined as a hidden damage.

[0048] As a further solution of the present invention, the method further includes a dynamic phase synchronization verification step: in the corrected hysteresis loop data, if the slope inflection points in the magnetization stage and the demagnetization stage are not completely aligned to the phase synchronization reference line, repeat steps S2 to S5 until the product error of the time difference between the magnetization stage and the demagnetization stage and the amplitude ratio

[0049] is less than 5%. The calculation method includes: performing moving window mean filtering on the magnetic field intensity data of the magnetization reference mark and the demagnetization reference mark, and taking 80% of the peak value after filtering as and the calculation reference value of

[0050] As a further solution of the present invention, the method further includes a damage verification step: when a hidden damage is detected, the conveyor belt is triggered to decelerate to 50% of the standard speed, the hysteresis loop data is re-acquired and secondary correction is performed. If the damage characteristics still exist after the secondary correction, a damage alarm signal is generated.

[0051] As a further solution of the present invention, the method for obtaining the hysteresis loop data includes: symmetrically arranging a magnetic induction probe array on both sides of the conveyor belt, collecting the magnetic field intensity distribution data in real time, and eliminating the environmental magnetic field interference through differential calculation.

[0052] See Figure 2 For the dynamic phase synchronization correction flow chart of the present invention, the figure shows the process of performing dynamic phase synchronization verification on the corrected hysteresis loop data. The user first views the corrected hysteresis loop data, and the data will be input into the slope inflection point recognition module and the time interval and amplitude ratio calculation module. After these two modules work together, the results will be input into the dynamic adjustment module; at the same time, the judgment step of analyzing whether the slope inflection points are aligned will evaluate the corrected hysteresis loop data. If they are not completely aligned, the re-identification step will be triggered and re-enter the calculation and / and the loop based on the amplitude ratio for adjustment until the result of analyzing whether the slope inflection points are aligned again is that the alignment is completed, that is, the alignment error is less than 5%. At this time, the new corrected data is output for the user to view, so as to ensure that the slope inflection points in the magnetization stage and the demagnetization stage can be accurately aligned. See Figure 3 , the hidden damage determination process provided by the present invention includes the following steps: first, input the corrected hysteresis loop data, and then calculate the local curvature to initially identify the abnormal changes in the hysteresis loop morphology; when the curvature is greater than the threshold, the system will extract the abnormal area and further calculate the magnetic field gradient to quantify the change rate of the magnetic field intensity along the spatial distribution; if the calculated gradient exceeds three times the historical benchmark, that is, it is determined that the gradient is greater than the historical benchmark × 3, then the hidden damage confirmation is executed; otherwise, if the curvature does not exceed the threshold or the gradient does not meet the determination conditions, it is respectively determined to be undamaged or the process is returned to continue the analysis to ensure that the damage recognition result has high accuracy and anti-interference ability. And as Figure 4As shown in the figure, the conveyor belt hidden damage detection system provided by the present invention includes a magnetic induction probe array, a hysteresis loop acquisition module, a data preprocessing module, a reference mark recognition module, a phase synchronization correction module, a damage determination module, and an alarm output module. Among them, the magnetic induction probe array is arranged on both sides of the conveyor belt for real-time acquisition of magnetic field intensity distribution data; the hysteresis loop acquisition module receives the probe signals and forms hysteresis loop data; the data preprocessing module performs operations such as denoising and moving average filtering on the original hysteresis loop; the reference mark recognition module determines the magnetization reference mark and the demagnetization reference mark by analyzing the slope mutation points in the magnetization stage and the demagnetization stage; the phase synchronization correction module dynamically adjusts the time axis and amplitude axis ratios of the hysteresis loop according to the time interval and amplitude ratio between the reference marks to achieve the correction of morphological distortion; the damage determination module identifies potential hidden damage according to the local curvature and magnetic field intensity gradient characteristics of the corrected hysteresis loop; the alarm output module generates an alarm signal after determining damage, realizing the high-robustness online monitoring ability of the system. See Figure 5 , the multi-dimensional damage analysis module of the present invention combines three feature dimensions of magnetic field intensity gradient direction, energy integral calculation, and time evolution analysis, and inputs them into the determination center for comprehensive evaluation. The magnetic field intensity gradient direction is used to reflect the spatial trend of local magnetic field changes in the material, the energy integral calculation is used to measure the energy difference of the hysteresis loop to identify the influence of load mutation, and the time evolution analysis is used to track the dynamic changes of characteristic points over time during magnetization and demagnetization processes. The determination center combines the above three data sources to achieve multi-angle cross-validation analysis of potential defects, and finally outputs the hidden damage result, significantly improving the determination reliability and result accuracy under complex working conditions.

[0053] Embodiment 1: In the current technology, the distortion of the hysteresis loop will bring inaccuracies to damage detection. Especially when the running speed of the conveyor belt changes, this distortion may mislead the determination of damage. Therefore, the specific steps adopted in this implementation description are as follows. Step 1: Obtain hysteresis loop data in real time. This step is achieved by arranging a magnetic induction sensor array on both sides of the conveyor belt to collect the hysteresis loop data during the operation of the conveyor belt in real time. These data include the correspondence between the magnetic field strength and the time series in the magnetization and demagnetization stages. The data acquisition system can automatically transmit these data to the processing system for further analysis. Step 2: Identify the slope inflection point in the magnetization stage. In the magnetization stage, the system calculates the first derivative of the curve of the magnetic field strength changing with time to identify the first mutation point of the slope. By this method, the magnetization reference mark is determined, avoiding the potential errors in the traditional method by using a fixed threshold or relying too much on hardware sensors. The key to this step lies in accurately calculating the change of the first derivative to ensure the more accurate identification of the reference mark. Step 3: Identify the slope inflection point in the demagnetization stage. Similarly, in the demagnetization stage, the system uses the curve of the magnetic field strength changing with time to calculate the mutation point of the slope. The identification method of this inflection point is similar to that in the magnetization stage to ensure the accurate determination of the demagnetization reference mark. This method avoids the errors that may be brought by using a fixed model in the traditional method and increases the adaptability of the system. Step 4: Calculation and dynamic adjustment of the time interval and amplitude ratio. Based on the time interval between the magnetization reference mark and the demagnetization reference mark, the amplitude ratio is calculated. The calculation formula of the amplitude ratio is the ratio of the magnetic field strength in the magnetization stage to that in the demagnetization stage. Through this ratio, the operation status of the conveyor belt is further analyzed. To cope with the interference caused by the speed fluctuation during the operation of the conveyor belt, the system dynamically adjusts the ratio of the time axis to the amplitude axis. This adjustment is carried out through the mapping relationship between the ratio and the preset synchronization reference line. Through this mechanism, the distortion caused by the change of the running speed can be eliminated in real time to ensure the accuracy of the data. Step 5: Correction of the hysteresis loop data and damage judgment. After the dynamic adjustment, the corrected hysteresis loop data is output. By analyzing these corrected data, it is determined whether there is hidden damage. The judgment of the damage location is based on the abnormal distortion characteristics of the magnetic field strength distribution, especially in the local curvature and the change rate of the gradient. Step 6: Dynamic phase synchronization verification. To further eliminate the influence of speed fluctuation, the system repeats the dynamic phase synchronization step according to the corrected data to ensure that the magnetization and demagnetization reference marks are completely aligned to the phase synchronization reference line. The completion degree of this verification step is high, ensuring the detection accuracy under complex working conditions.

[0054] To adapt to the effects of equipment aging and environmental changes, this embodiment introduces a dynamic adaptive update mechanism. By continuously monitoring the hysteresis loop data of the conveyor belt in a non-damaged state, the synchronization constant is recalculated regularly to compensate for the drift of the synchronization reference line caused by equipment aging. This design significantly improves the long-term stability and robustness of the system, ensuring high-efficiency hidden damage detection even under extreme working conditions. Compared with traditional methods, this embodiment no longer requires a complex multi-sensor array, but instead achieves high-efficiency and low-cost hidden damage detection through a single magnetic sensor combined with algorithm processing. The simplification of this hardware solution not only reduces the equipment investment cost but also improves the application flexibility of the system. In addition, to improve the credibility of detection, this embodiment comprehensively considers multiple factors such as the time evolution characteristics of the hysteresis loop, energy integral differences, and gradient direction distributions. Through this multi-dimensional cross-validation, not only surface cracks can be identified, but also hidden damages such as internal delamination can be accurately diagnosed.

[0055] Embodiment 2: In the amplitude ratio formula, is the magnetic field strength amplitude corresponding to the magnetization reference mark, defined as the amplitude when the magnetic field strength reaches the peak of its rising stage, is the magnetic field strength amplitude corresponding to the demagnetization reference mark, defined as the amplitude when the magnetic field strength drops to the lowest value in the demagnetization stage. Through the magnetic field strength amplitudes of these reference marks, the dynamic adjustment of the hysteresis loop can be carried out to eliminate the interference caused by speed fluctuations. The synchronization constant is a parameter determined under standard working conditions of the system, used to describe the relationship between the time interval between the magnetization and demagnetization stages and the magnetic field strength amplitude ratio. Its formula is: ,

[0056] where, is the magnetic field strength amplitude of the magnetization reference mark in a non-damaged state, is the magnetic field strength amplitude of the demagnetization reference mark in a non-damaged state, is the reference time interval between the two.

[0057] In this embodiment, by calculating the amplitude ratio of the magnetic field strength in the magnetization stage and the demagnetization stage, the health state of the conveyor belt can be effectively reflected. The amplitude ratio formula is: ,

[0058] where, ΔT is the time interval between the magnetization and demagnetization stages, is the time interval under the standard operating speed, is the preset synchronization constant.

[0059] Dynamic adjustment process: When is less than the preset threshold, the system will compress the time axis ratio and expand the amplitude axis ratio; when When it is greater than the preset threshold, the system will stretch the time axis ratio and shrink the amplitude axis ratio; this process ensures that the slope inflection points of the magnetization stage and the demagnetization stage are aligned to the phase synchronization reference line.

[0060] Through the above optimization steps, high-precision detection of hidden damage to the conveyor belt can be achieved, avoiding misjudgment caused by speed fluctuations or load changes, and ensuring the stability of the equipment after long-term use through an adaptive reference update mechanism. The specific implementation is as follows:

[0061] Step S1: Obtain the hysteresis loop data in real time. By arranging a magnetic induction probe array on both sides of the conveyor belt, the relationship between the magnetic field intensity and the time series during the operation of the conveyor belt is collected.

[0062] Step S2: In the magnetization stage, use the first derivative calculation to identify the slope inflection point and determine the position of the magnetization reference mark. Similarly, in the demagnetization stage, determine the demagnetization reference mark by a similar method.

[0063] Step S3: Calculate the amplitude ratio of the magnetic field intensity of the magnetization and demagnetization reference marks to determine the amplitude ratio , and compare it with the preset synchronization constant.

[0064] Step S4: If the amplitude ratio is less than the threshold, compress the time axis ratio and enlarge the amplitude axis ratio; if the amplitude ratio is greater than the threshold, stretch the time axis ratio and shrink the amplitude axis ratio to ensure that after the time axis and amplitude axis ratios of the hysteresis loop are adjusted, the magnetization and demagnetization reference marks are aligned to the phase synchronization reference line.

[0065] Step S5: Determine the position of the hidden damage based on the abnormal distortion characteristics of the magnetic field intensity distribution of the corrected hysteresis loop data. Further determine the damaged area through local curvature calculation and magnetic field intensity gradient analysis.

[0066] Step S6: In the corrected data, if the slope inflection points of the magnetization stage and the demagnetization stage are not completely aligned to the phase synchronization reference line, repeat steps S2 to S4 until the synchronization accuracy reaches the preset standard.

[0067] Step S7: Regularly collect the hysteresis loop data in the non-damaged state and update the synchronization constant to compensate for the influence of equipment aging or environmental factors on the hysteresis characteristics.

[0068] In the dynamic calibration mechanism of the present invention, effective calibration for speed fluctuations or load changes is achieved by dynamically adjusting the time axis ratio and the amplitude axis ratio. The system can achieve high robustness and low false positive rate under high-speed operation conditions; through endogenous phase difference perception and spatio-temporal dynamic adjustment, the high costs and synchronization problems of external sensors and complex data fusion models are avoided; by updating the synchronization constant in real time, the system can adapt to environmental factors such as equipment aging, temperature and humidity changes, ensuring long-term stable operation, which all belong to the extended implementation manners known to those of ordinary skill in the art.

[0069] Embodiment 3: In this embodiment, the preset phase synchronization constant can be obtained based on multiple hysteresis loop data acquisitions and analyses of the conveyor belt in a lossless state. Specifically, when the conveyor belt is in a healthy and undamaged state, hysteresis loop data at different operating speeds can be collected, and according to the formula:

[0070] ,

[0071] a reference phase synchronization constant is calculated. Among them, and respectively represent the magnetic field intensity amplitudes at the reference marks during the magnetization and demagnetization phases in the lossless state, and represents the time interval between these two reference marks in the lossless state. This reference value reflects the inherent relationship between the amplitude ratio and the time interval of the hysteresis loop under ideal conditions, and is used to correct the deviation caused by speed fluctuations during actual operation; the preset phase synchronization reference line essentially represents the target positions where the magnetization and demagnetization reference marks should align after the time axis and amplitude axis ratios are adjusted under the ideal uniform operation state. This reference line can be set according to statistical analysis of a large number of lossless conveyor belt samples or a theoretical model. For example, the reference line can be obtained by averaging the reference mark positions of the hysteresis loops of the lossless conveyor belt collected multiple times after synchronous calibration. And the threshold used to determine whether to adjust the time axis and amplitude axis ratios in step S4, as well as the threshold used to determine hidden damage in step S5, are all determined through experimental analysis and data statistics of a large number of conveyor belt samples in known states (including healthy and damaged to different degrees). The specific values of these thresholds will be adjusted according to factors such as the material, structure, and operating speed of the conveyor belt in the actual application scenario to achieve the best detection effect. For example, for the threshold to determine whether to adjust, a relatively small range can be set. When the actual amplitude ratio deviates from the synchronization constant When it exceeds this range, it is considered that correction is required. For the threshold value for determining damage, it needs to be set according to the characteristics of the hysteresis loop distortion caused by different types and degrees of damage to ensure that potential hidden damage can be effectively identified. And shows the characteristic amplitude ratio ( ) between the magnetization stage and the demagnetization stage in the hysteresis loop and the time interval ( ) between these two characteristic points, and there is a mapping relationship related to the conveyor belt running speed. In the ideal uniform speed state, this ratio should be close to the synchronous constant calculated from the lossless data. However, when the running speed of the conveyor belt fluctuates, will change accordingly, resulting in deviating from . The present invention makes the hysteresis loops collected at different speeds align to a unified benchmark by dynamically adjusting the ratio of the time axis and the amplitude axis of the hysteresis loop, thereby eliminating the interference caused by speed fluctuations and more accurately analyzing the magnetic characteristic changes caused by internal damage of the material. The target effect is to achieve decoupling of speed interference and highlight the abnormal distortion of the hysteresis loop caused by hidden damage; and the formula defines the phase synchronous constant , which is calculated from the reference data collected in the lossless state. and respectively represent the magnetic field strength amplitudes at the reference marks in the magnetization and demagnetization stages in the lossless state, is the time interval between these two reference marks. This constant can be regarded as an inherent magnetic characteristic parameter of the lossless conveyor belt in the ideal state. In the subsequent damage detection process, the amplitude ratio collected in real time can be compared with the calculated through the current time interval and the reference time interval , can be the average time interval in the lossless state), and the calculated is compared to determine whether correction is required, so as to provide a reliable reference value for subsequent dynamic correction through this formula. And it should be noted that and in the formula specifically refer to the magnetic field strength amplitudes at the slope inflection points identified by the first-order derivative calculation in the magnetization and demagnetization stages of the hysteresis loop. refers to the time interval between these two slope inflection points. can be understood as a reference time interval, such as the average time interval between these two inflection points in the lossless state, which all belong to the extended implementation manners known to those of ordinary skill in the art.

[0072] By introducing a dynamic correction mechanism based on the inherent phase attributes of the hysteresis loop, the present invention effectively eliminates the distortion of the hysteresis loop shape caused by the fluctuation of the conveyor belt running speed, enabling the data collected under complex working conditions to more accurately reflect the changes in the magnetic characteristics of the material itself. At the same time, by combining multi-dimensional data collaborative analysis methods such as local curvature calculation and magnetic field strength gradient analysis, it is possible to more precisely identify and locate hidden damages. For example, abnormal changes in local curvature can indicate the occurrence of damage, while sudden changes in magnetic field strength gradient help to determine the specific location of the damage. Moreover, the present invention mainly relies on in-depth analysis and processing of the data collected by the existing magnetic induction probes in the conveyor belt system, without the need to additionally add new hardware devices, thus achieving the detection of hidden damages of the conveyor belt without increasing the hardware cost. Through the dynamic reference adaptive calibration mechanism, the present invention can adapt to the operation of the conveyor belt under different working conditions and environmental conditions, and correct and analyze the hysteresis loop data in real time, thereby realizing all-weather and stable monitoring of the hidden damages of the conveyor belt.

[0073] The purpose of the moving window mean filtering mentioned in step S5 is to reduce the noise interference in the hysteresis loop data and make the subsequent damage feature analysis more accurate. For example, a moving window with a window size of 5 data points can be used to perform mean filtering on the hysteresis loop data. Specifically, when implemented, the window slides along the time series, calculates the average value of the 5 data points within the window each time, and takes this average value as the filtered magnetic field strength value at the current time point. The selection of the window size can be adjusted according to the characteristics of the noise in the actual application. The magnetic field strength gradient analysis identifies damages by calculating the rate of change of the magnetic field strength along the position on the hysteresis loop. Specifically, the difference in the magnetic field strength collected by two adjacent magnetic induction probes in the conveyor belt length direction can be calculated and divided by the distance between these two probes to obtain the magnetic field strength gradient. The historical reference value can be obtained by calculating the gradient of a known non-damaged conveyor belt area and taking its average value as the reference. When a significant deviation occurs between the gradient value calculated in real time and the historical reference value, there may be hidden damages. The synchronization accuracy threshold can be, for example, a product error less than 5% in the example. The setting of this threshold is based on a comprehensive consideration of the detection accuracy and system robustness. Specifically, when the product error of the time interval between the slope inflection points of the magnetization phase and the demagnetization phase is less than 5% after correction, it is considered that the synchronization accuracy meets the requirements and subsequent damage determination can be carried out. The specific value of this threshold can be adjusted according to the requirements of the actual application. For example, for scenarios with higher requirements for detection accuracy, the threshold can be set smaller, and all of these belong to the extended implementation methods known to those of ordinary skill in the art.

[0074] Example 4: In this example, the critical points where the slope first undergoes a sudden change in the magnetization and demagnetization phases are identified as reference marks. To improve the accuracy of inflection point identification under actual complex working conditions, a dynamic threshold adjustment mechanism is introduced in this example. Specifically, in the initial stage, the first derivative dropping below 20% of the peak value mentioned in the specification can be used as the preliminary inflection point identification threshold. However, considering the fluctuations in the conveyor belt running speed and the influence of sensor noise, this fixed threshold may not accurately identify the true inflection point in all cases. Therefore, in this example, an adaptive adjustment process is carried out after the preliminary identification of the inflection point. This process is based on the statistical analysis of historical data and the local characteristics of the current hysteresis loop. For example, a sliding window can be maintained to record the first derivative values of the identified magnetization and demagnetization inflection points in the recent period. Then, the mean and standard deviation of these historical first derivative values are calculated.

[0075] When identifying the inflection point of the current hysteresis loop, the first derivative value of the preliminarily identified inflection point is compared with the historical mean and standard deviation. If the current first derivative value deviates from the historical mean by more than a certain multiple of the standard deviation (e.g., 3 times the standard deviation), it is considered that the preliminarily identified inflection point may be a misjudgment caused by noise, and the identification threshold needs to be readjusted. The adjustment strategy can be: if the current first derivative value is much higher than the historical mean, the inflection point identification threshold is appropriately increased; if it is much lower than the historical mean, the threshold is appropriately decreased. The adjustment step size can be adaptively determined according to the deviation degree. Through this dynamic threshold adjustment mechanism, the robustness of inflection point identification can be improved, and the interference of speed fluctuations and noise can be reduced.

[0076] In the aspect of adaptively constructing and applying this reference line, in the initial stage, by analyzing the hysteresis loop data collected from a section of known non-damaged conveyor belt at different running speeds, the time intervals of its magnetization and demagnetization inflection points and the corresponding amplitude ratios are extracted. Then, by fitting these data points, an initial phase synchronization reference line can be obtained, and this reference line can be represented by a functional relationship . During the actual monitoring process, as the operating state of the conveyor belt changes, the initial reference line may no longer be fully applicable. Therefore, an adaptive update mechanism is introduced in this example. The hysteresis loop data in the current non-damaged state can be collected regularly (e.g., every once in a while or when a significant change in the operating state of the conveyor belt is detected), and the inflection point time intervals and amplitude ratios are recalculated, and the original reference line is updated using these new data points. The update method can be weighted average or re-function fitting.

[0077] When performing phase synchronization correction, for the currently collected hysteresis loop, first, the time intervals of its magnetization and demagnetization inflection points are identified and amplitude ratio . Then, substitute the time interval into the baseline function after adaptive update to obtain the expected amplitude ratio . By comparing the actual amplitude ratio and the expected amplitude ratio , the dynamic adjustment ratios for the time axis and amplitude axis of the hysteresis loop can be calculated. For example, the adjustment ratio of the time axis can be related to , and the adjustment ratio of the amplitude axis can be related to .

[0078] In addition to local curvature and magnetic field gradient analysis, the following features can also be introduced for comprehensive determination. For example, calculate the energy integral of the corrected hysteresis loop during magnetization and demagnetization processes. In the case of no damage, these two energy integrals should be close. When damage exists, it may lead to an increase in energy loss, thus increasing the difference between the two energy integrals. Specific frequency component analysis: Perform spectral analysis on the corrected magnetic field intensity signal to extract the amplitude and phase information of specific frequency components. Damage may cause abnormal signal changes within a specific frequency range. Deviation from historical data: Compare the current corrected hysteresis loop with the hysteresis loop in the historical non-damaged state and calculate the similarity or difference index between them. A significant deviation may indicate the existence of damage. At the same time, for each of the above features, corresponding thresholds can be set. When the value of one or more features exceeds its set threshold, it can be preliminarily determined that damage exists. To further improve the reliability of the determination, a weighted fusion method can be adopted. Different weights are assigned to each feature, and then based on the weighted comprehensive score, it is finally determined whether damage exists. The setting of the weights can be determined and adjusted based on a large amount of experimental data. Specifically, the initial setting of the weights can be based on the results of a large number of experimental analyses of different types and degrees of conveyor belt damage. For example, by collecting hysteresis loop data and extracting features from conveyor belts with known different types (such as cracks, wear, faults, etc.) and different severity levels (such as mild, moderate, severe) of damage, analyze the significance and reliability of features such as energy integral difference, amplitude of specific frequency components, deviation from historical data, local curvature, and magnetic field gradient under each damage type and degree. For features that show higher discrimination under a specific damage type or degree, higher initial weights can be assigned. And in practical applications, the rationality of the current weight setting can be evaluated by monitoring the performance of the system (such as false alarm rate and missed alarm rate). If it is found that a certain type of damage is often misjudged or missed, the weights of relevant features can be adjusted accordingly. For example, if the energy integral difference is easily interfered under a certain specific working condition, resulting in false alarms, its weight can be appropriately reduced. The adjustment methods can include regularly performing performance evaluations and manually adjusting the weights according to the evaluation results; or, an adaptive weight adjustment algorithm can also be introduced, such as a machine learning-based method, which automatically adjusts the weights of different features through continuous learning and optimization to achieve the best damage determination effect. These are all extended implementation methods known to those of ordinary skill in the art.

[0079] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above-described exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be embraced within the present invention.

Claims

1. A method for analyzing hidden damage of a conveyor belt using magnetic characteristics, characterized in that: The following steps are involved: Step S1, acquiring in real time the hysteresis loop data of the conveyor belt during operation, wherein the hysteresis loop data includes the corresponding relationship between the magnetic field intensity and the time series in the magnetization stage and the demagnetization stage; Step S2, identifying a first slope inflection point as a magnetization reference mark in the magnetization stage, wherein the first slope inflection point is a critical point where the slope first suddenly changes in a rising edge of the magnetic field intensity varying with time; Step S3, identifying a second slope inflection point as a demagnetization reference mark in the demagnetization stage, wherein the second slope inflection point is a critical point where the slope of the falling edge of the magnetic field intensity changing with time first undergoes a sudden change; Step S4, according to the time interval between the magnetized reference mark and the demagnetized reference mark , calculate the ratio of the magnetic field strength amplitudes of the two ,in is the magnetic field intensity amplitude corresponding to the magnetized reference mark, is the magnetic field intensity amplitude corresponding to the demagnetization reference mark, and the amplitude ratio satisfies the following relationship: ， in, is the preset phase synchronization constant, It is the reference time interval of the conveyor belt at the standard operating speed; Step S5, based on the amplitude ratio The mapping relationship with the preset phase synchronization baseline dynamically adjusts the time axis ratio and the amplitude axis ratio of the hysteresis loop so that the slope inflection points of the adjusted magnetization stage and demagnetization stage are aligned to the phase synchronization baseline; Step S6, outputting the corrected hysteresis loop data, and determining the hidden damage position of the conveyor belt according to the abnormal distortion characteristics of the magnetic field intensity distribution in the corrected data.

2. The method for analyzing hidden damage of a conveyor belt using magnetic characteristics according to claim 1, characterized in that: The method for identifying the first slope inflection point in step S2 includes: calculating the first-order derivative of the curve of magnetic field intensity variation with time in the magnetization stage, and defining the position where the derivative first drops below 20% of the peak value as the first slope inflection point.

3. The method for analyzing hidden damage of a conveyor belt using magnetic characteristics according to claim 2, characterized in that: The dynamic adjustment in step S5 includes: When it is less than a preset threshold, the time axis ratio of the hysteresis loop is compressed and the amplitude axis ratio is enlarged; when When the time difference between the magnetized reference mark and the demagnetized reference mark is greater than a preset threshold, the time axis ratio is stretched and the amplitude axis ratio is reduced until the time difference between the magnetized reference mark and the demagnetized reference mark is Ratio to Amplitude The product of reaches the preset synchronization constant.

4. The method for analyzing hidden damage of a conveyor belt using magnetic characteristics according to claim 1, characterized in that: The method for setting the phase synchronization reference line includes: collecting the temporal and spatial distribution characteristics of the magnetization reference mark and the demagnetization reference mark of the hysteresis loop when the conveyor belt is running at the standard speed when the conveyor belt is not damaged, and calculating the synchronization constant by the following formula : ， in, is the magnetic field intensity amplitude of the magnetized reference mark in the undamaged state, is the magnetic field intensity amplitude of the demagnetized reference mark in an undamaged state, is the base time interval for both.

5. The method for analyzing hidden damage of a conveyor belt using magnetic characteristics according to claim 1, characterized in that: The abnormal distortion feature determination method in step S6 includes: calculating the local curvature of the corrected hysteresis loop, marking the continuous area whose curvature exceeds a preset threshold as a potential damage area, and performing a magnetic field intensity gradient analysis on the area. If the gradient change rate exceeds 3 times of the historical baseline value, it is determined to be a hidden damage.

6. The method for analyzing hidden damage of a conveyor belt using magnetic characteristics according to claim 5, characterized in that: The method further comprises a dynamic phase synchronization verification step: in the corrected hysteresis loop data, if the slope inflection points of the magnetization phase and the demagnetization phase are not completely aligned to the phase synchronization reference line, then steps S2 to S5 are repeatedly executed until the time difference between the two is equal to or greater than 100. Ratio to Amplitude The product error is less than 5%.

7. The method for analyzing hidden damage of a conveyor belt using magnetic characteristics according to claim 6, characterized in that: The amplitude ratio in step S4 The calculation method includes: performing sliding window mean filtering on the magnetic field intensity data of the magnetized reference mark and the demagnetized reference mark, and taking 80% of the peak value after filtering as and The calculation base value of .

8. The method for analyzing hidden damage of a conveyor belt using magnetic characteristics according to claim 1, characterized in that: The method also includes a damage verification step: when hidden damage is detected, the conveyor belt is triggered to slow down to 50% of the standard speed, the hysteresis loop data is recollected and a secondary correction is performed, and if the damage feature still exists after the secondary correction, a damage alarm signal is generated.

9. The method for analyzing hidden damage of a conveyor belt using magnetic characteristics according to claim 1, characterized in that: The method for acquiring hysteresis loop data includes: symmetrically arranging magnetic induction probe arrays on both sides of the conveyor belt, collecting magnetic field intensity distribution data in real time, and eliminating environmental magnetic field interference through differential calculation.

Images