Method for monitoring the state of a cross-sea stockpile line transport belt

Abstract

The application discloses a state monitoring method based on a cross-sea stockpile line transportation belt, relates to the technical field of cross-sea stockpile line transportation belt state monitoring, and realizes comprehensive monitoring through four steps. In step one, a belt empty load reference feature is collected, foreign matter and tearing are monitored through AI visual recognition, in step two, when visual failure occurs, vibration and temperature double-dimension monitoring are used to cooperatively identify abnormalities, in step three, the risk of foreign matter evolving into tearing and tearing causes are evaluated, and in step four, multi-modal monitoring reports are integrated to form a closed-loop operation and maintenance. Through multi-dimension data fusion of vision, vibration and temperature, combined with reference feature comparison and dynamic verification, rod-shaped foreign matter, net-shaped foreign matter and tearing can be accurately identified, the problem of visual failure caused by material shielding is effectively solved, the accuracy of abnormal identification is improved, closed-loop management from monitoring to disposal is realized, the safe and stable operation of the cross-sea stockpile line transportation belt is ensured, operation and maintenance costs are reduced, and transportation efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of state monitoring of cross-sea stockpile line conveying belts, in particular to a state monitoring method based on cross-sea stockpile line conveying belts. BACKGROUND

[0002] The cross-sea stockpile line conveying belt is a key equipment in logistics scenes such as ports, and the traditional monitoring relies on manual inspection, which is low in efficiency and high in missed detection rate. Complex working conditions such as material shielding often lead to visual monitoring failure, and faults are easily caused by foreign matter blockage and tearing, which affects transportation. Therefore, a precise and multi-dimensional monitoring method is urgently needed.

[0003] The existing or traditional state monitoring method based on cross-sea stockpile line conveying belts has the following technical problems: 1. The existing state monitoring method based on cross-sea stockpile line conveying belts lacks a unified state monitoring reference feature setting, and the reference image, average vibration signal and average temperature in the empty state are not collected by the system, which will lead to a lack of unified reference standard in the monitoring process, and the monitoring data of different time periods and different equipment cannot be effectively compared, which is prone to misjudgment due to environmental interference or equipment error, and it is difficult to distinguish between normal operation fluctuations and abnormal states.

[0004] 2. The traditional state monitoring method based on cross-sea stockpile line conveying belts lacks hierarchical processing of visual monitoring results, and the monitoring positions of double-dimensional monitoring key and non-key foreign matters and tearing are not distinguished, which will lead to the use of the same monitoring intensity for all regions when the vision is not failed, causing resource waste, and when the vision is failed, the regions that need to be monitored cannot be accurately locked, which is prone to missed detection, especially in the working condition of frequent material shielding, the abnormal identification efficiency is greatly reduced.

[0005] 3. The traditional state monitoring method based on cross-sea stockpile line conveying belts lacks a double-dimensional cooperative monitoring mechanism of vibration and temperature, and only relies on single visual monitoring or independent vibration or temperature monitoring, which will lead to the inability to identify abnormalities through the characteristic matching of vibration signals and temperature changes when the vision is failed due to material shielding, and it is difficult to judge the rod-shaped objects, net-shaped objects or tearing state of the shielded area, increasing the safety hidden danger caused by shielding.

[0006] 4. The traditional state monitoring method based on cross-sea stockpile line conveying belts lacks a warning mechanism for the evolution risk of foreign matter to tearing, and the dynamic changes of foreign matter vibration and temperature characteristics are not tracked, which will lead to the inability to identify the evolution process of "blockage-friction aggravation-tearing" in advance, and only the tearing can be handled after it occurs, increasing the equipment downtime and maintenance cost, and even causing more serious chain faults. SUMMARY

[0007] The purpose of the present application is to provide a state monitoring method based on cross-sea stockpile line conveying belts, which solves the problems in the background art.

[0008] To solve the above technical problems, the application adopts the following technical solutions: The application provides a state monitoring method based on a cross-sea stockpile line transport belt, comprising: step one, AI visual preliminary monitoring: collecting state monitoring reference features of each belt under an empty state, including a fiber stethoscopic vibration reference, and performing transport foreign matter monitoring and tearing monitoring on each belt through AI visual recognition combined with the fiber vibration reference.

[0009] Step two, double-dimension fusion monitoring: when material shielding leads to visual failure, performing transport foreign matter monitoring and tearing monitoring on the belt through double-dimension methods of fiber stethoscopic vibration monitoring and temperature monitoring, and comprehensively evaluating the corresponding state of the belt.

[0010] Step three, state correlation evaluation: according to the double-dimension monitoring collaborative recognition result, taking the fiber stethoscopic vibration data as the core, warning the evolution risk of foreign matter to tearing within a set time period, and evaluating the cause of belt tearing.

[0011] Step four, multi-modal fusion: integrating visual, fiber stethoscopic vibration, and temperature monitoring reports, and then forming a closed-loop operation and maintenance process.

[0012] The application has the following beneficial effects: 1. The state monitoring method based on the cross-sea stockpile line transport belt provided by the application records the reference image, average vibration signal, and average temperature without foreign matter and tearing through the deployment of high-frame-rate industrial cameras, distributed fiber stethoscopic sensors, thermal imaging cameras, and other equipment during the process of collecting the state monitoring reference features of each belt under an empty state, thereby providing a unified reference standard for subsequent monitoring, which is conducive to eliminating the recognition deviation caused by the lack of reference and ensuring the accuracy of abnormal judgment.

[0013] 2. In the AI visual preliminary monitoring process of the application, the real-time image is compared with the reference template to distinguish between double-dimension monitoring key and non-key foreign matter and tearing monitoring positions, and the non-key positions are pushed to manual monitoring and the key positions are started for double-dimension monitoring, which is conducive to realizing the reasonable allocation of monitoring resources, avoiding the consumption of resources by invalid monitoring, and reducing the risk of missed judgment when the vision fails.

[0014] 3. In the double-dimension fusion monitoring process of the application, when the vision fails, the vibration signal monitoring and temperature value monitoring work cooperatively to match and judge the characteristics of rod-shaped objects, net-shaped objects, and tearing, which is conducive to making up for the limitations of single visual monitoring and accurately identifying abnormalities under material shielding working conditions.

[0015] 4、The embodiment of the present application in the state correlation evaluation process, by tracking the foreign matter vibration and temperature characteristic change, backtracking the history data before tearing, early warning the evolution risk of foreign matter to tearing and clear tearing cause, is beneficial to intervene in potential risk point in advance, reduce the shutdown loss caused by tearing, and provide basis for targeted maintenance.

[0016] 5、The embodiment of the present application in the multi-modal fusion process, integrates visual, vibration and temperature monitoring reports, removes single modal false alarm through cross validation, forms a comprehensive report containing abnormal type, location, evolution stage and cause, promotes operation and maintenance scheme development, effect tracking and benchmark data updating, is beneficial to build the complete process of "monitoring-analysis-disposal-closed loop", improve operation efficiency, and guarantee the continuous and stable operation of the cross-sea stacking line transport belt. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of these drawings.

[0018] Figure 1 The embodiment of the present application is a flowchart of the steps. DETAILED DESCRIPTION

[0019] The technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0020] Please refer to Figure 1 The present application provides a state monitoring method based on cross-sea stacking line transport belt, which comprises the following steps: step one, AI visual preliminary monitoring: collecting state monitoring benchmark characteristics of each belt under no-load state, including fiber auscultation vibration benchmark, and monitoring and tearing of each belt by AI visual identification combined with fiber vibration benchmark.

[0021] In a specific embodiment, the state monitoring reference features of each belt in an idle state are collected, and the specific process is as follows: in the idle state of the belt without transporting materials and without running, high-frame-rate and high-pixel industrial cameras and dust-penetrating cameras are deployed above the head and tail of each belt and the side plate of the material falling port sealing device, and distributed fiber stethoscopic sensors are fixed along the belt steel frame, and a four-channel stethoscopic fiber main machine is placed in the SR102 head duty room, and a thermal imaging camera is installed on the belt below the roller device bridge and is aligned with the belt bearing surface.

[0022] The state monitoring reference features include visual reference, vibration reference, and temperature reference, wherein the vibration reference includes the mechanical noise characteristics of the roller, and the temperature reference includes the normal working temperature of the roller bearing, which is collected by the temperature sensor installed on the roller bearing seat, and then calculated by the mean value to serve as the temperature reference. The high-frame-rate camera and the dust-penetrating camera are started to scan the full surface of each belt, and the reference images without foreign matter and tearing are recorded, including the belt edge profile, the roller position, and the surface flatness, which are stored in the database as the visual comparison template.

[0023] The materials screened without foreign matter are transported through each type of belt for transportation testing, the vibration signals at each collection time point at the corresponding feeding port of each belt are collected by the distributed fiber stethoscopic sensors, the average vibration signals at each collection time point at the corresponding feeding port of each belt are obtained by mean value calculation, the average vibration signals include the average vibration frequency of the roller rotation and the average vibration frequency of the belt, and thus the roller rotation frequency threshold range and the belt vibration frequency range of each belt are set as the average vibration frequency of the roller rotation ± 30% and the average vibration frequency of the belt ± 30%, respectively, and the belt surface temperature and the roller temperature at each collection time point during the transportation testing of each belt are recorded by the temperature sensor and the thermal imaging camera, and the average belt surface temperature and the average roller temperature corresponding to the transportation testing of each belt are obtained by mean value calculation.

[0024] It should be noted that in the belt transportation scene, the "feeding port of each belt" refers to the entrance position of the material into each belt. Specifically, the core function of the belt conveyor is to transport materials from one place to another, and the "feeding port" is the starting point of the material into the current belt. When the material enters the belt, it will have a direct effect on the belt and the rollers near the feeding port, such as material impact on the belt and force change of the driven roller, etc., so the vibration signal at the feeding port can reflect the running state of the belt in the key link of receiving materials and starting transportation.

[0025] It should also be noted that the high-frame-rate in the industrial camera refers to 30 frames per second and above, which can capture the details of the belt running at a high speed, and the high-pixel refers to 2 million pixels and above, which can clearly present the surface and edge features of the belt.

[0026] In a specific embodiment, the AI visual recognition method is used to monitor the transported foreign matter of each belt, and the specific process is as follows: during the belt material transportation process, the industrial cameras and dust-penetrating cameras arranged above the corresponding head and tail of each belt and the material falling port collect the belt surface images at the preset collection time interval, and the light supplement device is started, and the self-cleaning air device is used. The AI visual recognition algorithm first calls the visual comparison template in the database, compares the real-time collected image with the reference template, and if there is an abnormality after comparing the image at a certain collection time point with the reference template image, the image at the collection time point is recorded as a suspected foreign matter image, and the position is recorded as a double-dimension monitoring non-key foreign matter monitoring position. At the same time, the position is labeled on the cloud platform and pushed to the mobile APP to remind the operation and maintenance personnel to conduct manual monitoring.

[0027] If there is no abnormality after comparing the image at a certain collection time point with the reference template image, it is determined that the actual position of the image at the collection time point is blocked by the material, resulting in visual failure, and the actual position of the image at the collection time point is recorded as a double-dimension monitoring key foreign matter monitoring position. Then, the double-dimension monitoring method is used to monitor the double-dimension monitoring key monitoring position.

[0028] It should be noted that while the AI visual recognition method is used to monitor the transported foreign matter of each belt, the deviation of each belt is also monitored. By comparing the deviation of the belt edge in the real-time image with the reference contour, if the deviation exceeds the set threshold, it is determined that the belt is deviated, and a warning notification is triggered.

[0029] It should be further noted that during the process of pushing to the mobile APP to remind the operation and maintenance personnel to conduct manual monitoring, the fiber vibration sound data of the position is also synchronized to assist manual labeling and judgment. For example, when the AI visual recognition detects that the image of a certain area is abnormal compared with the reference template, such as an irregular shadow appears on the belt surface, and it cannot be determined whether it is a foreign matter or a material accumulation, the system marks the position as a double-dimension monitoring non-key foreign matter monitoring position and pushes it to the mobile APP. At this time, the fiber vibration sound data of the position is synchronized, and if the sound is a regular high-frequency "click" impact sound, which meets the collision characteristics of the rod-shaped object and the belt, the operation and maintenance personnel can determine it as a rod-shaped object and complete the labeling. If the sound is only a chaotic low-frequency friction sound without obvious foreign matter collision characteristics, it can be determined as material accumulation rather than foreign matter. The sound data supplements the ambiguity of visual information, thereby improving the accuracy of manual labeling.

[0030] It also needs to be explained that, assuming the reference template is the surface image of the belt when it is empty, there are no foreign objects, the edge is flat, and the roller position is clear. When a long strip-shaped object appears on the surface of the belt in the real-time image at a certain collection time point, or the edge appears a clear crack, it belongs to "abnormality". The system marks the two-dimensional monitoring non-key foreign object monitoring position, pushes the manual confirmation, and if the real-time image is completely covered by the material, the visual features of the covered area are highly similar to the no-foreign-object area in the corresponding position of the reference template, and it cannot be identified whether there is a foreign object or a tear, which belongs to "no abnormality". It is determined that the material shielding leads to visual failure, marked as a two-dimensional monitoring key foreign object monitoring position, and the vibration and temperature monitoring is started.

[0031] In a specific embodiment, the tearing of each belt is monitored by AI visual recognition, and the specific process is as follows: when the tearing is monitored by AI visual recognition, the industrial camera shoots the edges and surfaces of each belt in real time according to the preset collection time interval, the AI algorithm calls the edge profile and surface flatness data of the no-tear state in the state monitoring reference feature, compares the edges and surfaces of each belt shot in real time with the edge profile and surface flatness data of the no-tear state in the state monitoring reference feature at the pixel level, detects whether there is a burr that deviates from the reference profile by more than a set burr length threshold, and if the detection result shows a burr that deviates from the reference profile by more than a set burr length threshold, the actual position corresponding to the image that deviates from the reference profile is recorded as a two-dimensional monitoring non-key tearing monitoring position, and the position is labeled on the cloud platform and pushed to the mobile APP to remind the operation and maintenance personnel to perform manual monitoring. If the detection result does not show a burr that deviates from the reference profile by more than a set burr length threshold, it is determined that material shielding leads to visual failure, and the actual position corresponding to the image that does not deviate from the reference profile is recorded as a two-dimensional monitoring key tearing monitoring position, and then the two-dimensional monitoring key tearing monitoring position is monitored by two-dimensional monitoring.

[0032] It needs to be explained that while the tearing of each belt is monitored by AI visual recognition, the running deviation of each belt is monitored by comparing the edge deviation distance at the pixel level, and if the edge deviation distance exceeds the reference range, a warning is triggered.

[0033] It also needs to be explained that in the process of pushing to the mobile APP to remind the operation and maintenance personnel to conduct artificial monitoring, the vibration soundprint features collected by the optical fiber auscultation sensor need to be combined to achieve the purpose of improving the accuracy of artificial labeling, for example: when the AI visual recognition finds that there is a fuzzy abnormal contour on the belt edge, such as a small tear or attached blocky material, if the vibration soundprint collected by the optical fiber auscultation sensor is combined when the vibration soundprint presents high-frequency, intermittent "bump" sound, which is consistent with the friction characteristics of blocky material and the belt, then the operation and maintenance personnel can determine that it is foreign matter attachment, if the vibration soundprint presents low-frequency, continuous "scraping" sound, which is consistent with the friction characteristics of the tear and the roller, then it can be determined as early tearing, through the complement of the soundprint features and the visual image, the misjudgment caused by relying only on the image blur is avoided, thereby improving the accuracy of artificial labeling.

[0034] In the AI visual preliminary monitoring process, the real-time image is compared with the reference template to distinguish between double-dimension monitoring key and non-key foreign matter and tear monitoring positions, artificial monitoring is pushed to non-key positions and double-dimension monitoring is started on key positions, which is beneficial to realize reasonable allocation of monitoring resources, avoid invalid monitoring to consume resources, and reduce the risk of missed judgment when the vision fails.

[0035] Step two, double-dimension fusion monitoring: when the material shielding causes the vision to fail, the double-dimension method of optical fiber auscultation vibration monitoring and temperature monitoring is used to monitor the transported foreign matter and tear of the belt, and the corresponding state of the belt is comprehensively evaluated.

[0036] In one specific embodiment, the transported foreign matter and tear of the belt are monitored by the optical fiber auscultation vibration monitoring, and the specific process is as follows: the double-dimension monitoring method is vibration signal monitoring and temperature value monitoring, when the material shielding causes the vision to fail, the distributed optical fiber auscultation sensor starts to operate first, according to the belt section where the double-dimension monitoring non-key foreign matter monitoring position is located, the sensor collects the roller rotation frequency and the belt vibration frequency at the roller rotation frequency threshold and the belt vibration frequency recorded when the material without foreign matter is transported, and compares them, if high-frequency pulse vibration is detected, and the roller rotation frequency and the belt vibration frequency are not within the roller rotation frequency threshold range and the belt vibration frequency range, then it is determined as a blocked rod-shaped object.

[0037] If low-frequency continuous vibration occurs, the roller rotation frequency and the belt vibration frequency are not within the roller rotation frequency threshold range and the belt vibration frequency range, and the amplitude is higher than the average signal by a set percentage, and the frequency spectrum presents a multi-frequency superposition feature, it is determined that the net is blocked. If low-frequency strong vibration is captured, the roller rotation frequency and the belt vibration frequency are not within the roller rotation frequency threshold range and the belt vibration frequency range, the amplitude is higher than the average signal by a set percentage, and the duration exceeds a set time, it is determined that the tear is blocked. Then, the temperature value monitoring method is used for transport foreign object monitoring and tear monitoring.

[0038] In a specific embodiment, the belt is monitored for transport foreign objects and tears by temperature monitoring. The specific process is as follows: The temperature value monitoring method works synchronously with the thermal imaging camera installed on the roller device bridge under the belt. The temperature of the blocked area is recorded in real time according to the preset collection time interval, and compared with the average belt surface temperature and the average roller temperature obtained by transport testing. When monitoring the double-dimension monitoring key foreign object monitoring position, the temperature at the corresponding position changes synchronously with the vibration pulse, showing instantaneous rise and fall, and the temperature difference fluctuation conforms to the preset pulse temperature difference threshold based on the average belt surface temperature and the average roller temperature, i.e. it matches the vibration characteristics of the rod-shaped object. It is determined that there is a rod-shaped object blocked by material at the double-dimension monitoring key tear monitoring position. If the temperature shows a stepwise increase, the cumulative temperature difference per unit time is within the preset stepwise temperature difference range with reference to the reference temperature, and the overall temperature deviates from the average belt surface temperature and the average roller temperature to the extent that the net friction characteristics are consistent, it is determined that there is a rod-shaped object blocked by material at the double-dimension monitoring key tear monitoring position.

[0039] If the temperature at the double-dimension monitoring key tear monitoring position shows a linear rise, the high-temperature area forms a continuous strip with the belt running and there is no pulse fluctuation, and the temperature difference change amplitude conforms to the preset tear characteristic temperature difference range based on the reference temperature, it is determined that there is a tear blocked by material at the double-dimension monitoring key tear monitoring position.

[0040] It should be noted that "showing instantaneous rise and fall" means that the temperature changes rapidly synchronously with the vibration pulse. When the rod-shaped object intermittently collides with the roller and the belt, heat is generated by friction at each collision moment, the temperature rises suddenly, and the heat is gradually dissipated after the collision, the temperature slowly falls, forming a pulse change from sudden rise to slow fall. The "preset pulse temperature difference threshold" is a reasonable fluctuation range based on the average belt surface temperature and the average roller temperature, used to determine whether the instantaneous change conforms to the rod-shaped object characteristics.

[0041] The "stepwise increase" refers to the fact that the temperature of the mesh does not increase continuously and uniformly, but increases by a certain amplitude every certain period of time, showing a stepwise trend of stability to small jumps to stability again, because the friction strength between the mesh and the belt is relatively stable but the heat is continuously accumulated. The "pre-set stepwise temperature difference range" is the temperature increase amplitude per unit time and the overall temperature difference upper limit set based on the reference temperature, which is used to match the friction characteristics of the mesh.

[0042] It should be further noted that, assuming that the belt is transporting materials, a certain area is blocked by the materials, and the reference average belt surface temperature is 25°C. If the thermal imaging shows that the temperature of the area increases from 25°C to 30°C and then falls again with vibration, which meets the pulse temperature difference threshold, it is determined that there is a rod-shaped object, if the temperature increases by 3°C every 30 seconds, and the overall temperature reaches 37°C, which meets the stepwise range, it is determined that there is a mesh, if the temperature increases by 10°C every minute, forming a continuous strip-shaped high-temperature zone, it is determined that there is a tear, and the examples are only exemplary, and the examples are not the only limitation.

[0043] In a specific embodiment, the comprehensive evaluation of the state of the belt is as follows: if the roller rotation frequency and the belt vibration frequency do not belong to the roller rotation frequency threshold range and the belt vibration frequency range, and the difference between the corresponding temperature change and the average temperature meets the temperature matching relationship, the abnormal type is confirmed, and then the position of the optical fiber sensor is located to determine the specific area of the abnormality on the belt, if the roller rotation frequency and the belt vibration frequency do not belong to the roller rotation frequency threshold range and the belt vibration frequency range, but the temperature has no corresponding difference with the average temperature, it is determined that the material is unevenly disturbed, and no alarm is triggered, if the temperature deviates from the average temperature, but the vibration signal has no corresponding characteristics, it is determined that the material itself has temperature difference, and no alarm is triggered.

[0044] In the dual-dimension fusion monitoring process, when the vision fails, the vibration signal monitoring and temperature value monitoring work cooperatively to match and judge the characteristics of rod-shaped objects, mesh objects and tears, which is beneficial to overcome the limitations of single visual monitoring and can still accurately identify abnormalities under material blocking conditions.

[0045] Step three, state correlation evaluation: based on the dual-dimension monitoring and cooperative identification results, the optical fiber auscultation vibration data is taken as the core to warn the evolution risk of foreign objects to tears in a set time period, and the causes of belt tearing are evaluated.

[0046] In a specific embodiment, the process of early warning the evolution risk of the foreign matter to tear in a set time period is as follows: when early warning the evolution risk of the foreign matter to tear in a set time period, by tracking the changes of vibration and temperature characteristics of the identified foreign matter and comparing with the state monitoring benchmark characteristics of the transportation test, the vibration soundprint signal of the foreign matter captured by the fiber stethoscopic sensor is monitored in real time to determine whether the frequency deviation further expands from the threshold range of the roller rotation frequency and the range of the belt vibration frequency, whether the amplitude continues to rise, and whether the temperature rise rate changes from pulse and ladder to linear growth. If all the above characteristics appear in the set time period, and the low-frequency strong vibration trend characteristic of tearing appears in the vibration spectrum, it is determined that the foreign matter has entered the evolution stage from blocking to friction aggravation to imminent tearing, and an upgraded warning is triggered to prompt that the belt will tear.

[0047] It should be noted that, assuming that the threshold range of the roller rotation frequency set by the transportation test is 70-130 Hz, and the threshold range of the belt vibration frequency is 35-65 Hz, if the initial identified rod-shaped object vibration has a roller rotation frequency of 135 Hz and a belt vibration frequency of 70 Hz, and the subsequent 30 seconds tracking finds that the roller rotation frequency rises to 150 Hz and the belt vibration frequency rises to 85 Hz, i.e. the frequency deviation "further expands", at the same time, the amplitude increases from the initial 0.1 mm to 0.3 mm, the temperature change changes from the original pulse to linear growth, and the low-frequency strong vibration of 500-1.5 kHz appears in the vibration spectrum, at this time it is determined that the rod-shaped object has entered the friction aggravation stage from blocking, and is about to cause tearing, and an upgraded warning is triggered immediately.

[0048] It should be noted that "the temperature rise rate changes from pulse and ladder to linear growth" means that: the temperature change of the foreign matter in the initial stage conforms to its own characteristics, the rod-shaped object is pulse, i.e. it rises and falls instantaneously with the vibration pulse and then falls back; the mesh-shaped object is ladder, i.e. it rises by a certain amplitude every certain time and then stabilizes for a short time, and when the foreign matter evolves to tear, the mode of temperature rise changes from the original intermittent and phased rise to continuous and uniform rise, i.e. the temperature rise amplitude in unit time is basically consistent, without obvious falling back or pause, for example, if the initial identification is a rod-shaped object, the temperature change is a pulse from rising 5°C when vibrating to falling back to the benchmark when vibrating stops, which is pulse, when the rod-shaped object blocks and aggravates, the friction changes from intermittent to continuous, and the temperature changes from rising and falling to stably rising 8°C per minute without falling back, which is linear growth. Similarly, if the mesh-shaped object changes from the ladder of rising 3°C every 30 seconds to the linear growth of uniformly rising 6°C per minute, it means that the friction has entered the continuous aggravation state, which is a signal of evolution to tear.

[0049] In a specific embodiment, the evaluation of the cause of the belt tear is specifically as follows: according to the two-dimensional monitoring data at the time of the tear, the two-dimensional monitoring data including vibration signal data and temperature value data, the historical record of the tear setting time period is traced back, if there is vibration signal conforming to the foreign matter characteristics in the historical record of the tear setting time period, and the temperature anomaly of the corresponding position matches the foreign matter characteristics, it is determined that the tear is caused by foreign matter blocking;

[0050] If there is no foreign matter related vibration and temperature anomaly in the historical record of the tear setting time period, only the vibration and temperature characteristics specific to the tear exist, it is determined that the tear is caused by non-foreign matter factors.

[0051] It should be noted that the two-dimensional monitoring data at the time of the tear is based on the vibration soundprint data recorded by the optical fiber auscultation sensor to evaluate the cause of the belt tear.

[0052] In the state correlation evaluation process, the evolution risk of foreign matter to tear is warned and the cause of the tear is determined by tracking the change of foreign matter vibration and temperature characteristics and tracing back the historical data before the tear, which is beneficial to intervene in the potential risk point in advance, reduce the shutdown loss caused by the tear, and provide a basis for targeted maintenance.

[0053] Step four, multi-modal fusion: integrate visual, fiber auscultation vibration and temperature monitoring reports, and then form a closed-loop operation and maintenance process.

[0054] In a specific embodiment, the integration of visual, vibration and temperature monitoring reports is specifically as follows: according to the reference features and real-time recognition results of visual monitoring, the abnormal frequency spectrum and frequency deviation data of vibration monitoring, the abnormal temperature rising trend and reference temperature difference information of temperature monitoring, combined with foreign matter evolution risk warning and tear cause evaluation conclusion, single modal false alarm is eliminated through cross validation, comprehensive report containing abnormal type, accurate position, evolution stage and cause is formed, and the closed-loop process from monitoring to solving is completed based on the basis of promoting the development of disposal scheme, effect tracking and reference data dynamic updating of operation and maintenance links.

[0055] In the multi-modal fusion process of the embodiment of the application, the visual, vibration and temperature monitoring reports are integrated, the single modal false alarm is eliminated through cross validation, the comprehensive report containing the abnormal type, position, evolution stage and cause is formed, the development of operation and maintenance scheme, effect tracking and reference data updating are promoted, which is beneficial to build the complete process of "monitoring-analysis-disposal-closed loop", improve the operation and maintenance efficiency, and ensure the continuous and stable operation of the cross-sea stacking line transport belt.

[0056] The state monitoring method based on the cross-sea stockpile line transportation belt provided by the embodiment of the application, through the process of collecting the state monitoring reference characteristics of each belt under the no-load state, through the deployment of high frame rate industrial cameras, distributed optical fiber auscultation sensors, thermal imaging cameras and other equipment, the reference images without foreign matter and tearing, the average vibration signal and the average temperature are recorded, a unified reference standard is provided for subsequent monitoring, which is beneficial to eliminate the identification deviation caused by the lack of reference, and ensures the accuracy of the abnormal judgment.

[0057] The above is only an example and description of the concept of the present application. Those skilled in the art can make various modifications or supplements to the described specific embodiments or use similar ways to replace, as long as the modifications or supplements do not deviate from the concept of the present application or exceed the scope defined in the specification, which shall belong to the protection scope of the present application.

Claims

1. A method for condition monitoring of a cross-sea stockpile line transport belt, characterized in that, The application relates to a double-dimension fusion monitoring method for belt conveying, and belongs to the technical field of belt conveying. Step one, AI visual preliminary monitoring: collecting state monitoring reference features of each belt in an empty state, including fiber auscultation vibration, and performing conveying foreign matter monitoring and tearing monitoring on each belt through AI visual identification combined with fiber auscultation vibration; The specific process of collecting the state monitoring reference features of each belt in the empty state is as follows: In the empty state of the belt without conveying materials and without running, first, the distributed fiber auscultation sensor is fixed along the belt steel frame, the four-channel auscultation fiber host is arranged in the SR102 machine head value room, and then a high-frame-rate and high-pixel industrial camera and a dust-penetrating camera are arranged above the machine head and the tail of each belt and on the side plate of the material falling port sealing device, and a thermal imaging camera is installed on the bridge of the roller device below the belt and is aligned with the belt bearing surface; The state monitoring reference features include visual reference, vibration reference and temperature reference, wherein the vibration reference contains mechanical noise features of the roller, and the temperature reference contains the normal working temperature of the roller bearing, the normal working temperature of the roller bearing is collected through the temperature sensor arranged on the roller bearing seat, and then the temperature reference is obtained through mean value calculation, the high-frame-rate camera and the dust-penetrating camera are started, the full surface of each belt is scanned, the reference images without foreign matters and tearing are recorded, including the belt edge profile, the roller position and the surface flatness, and the reference images are stored in the database as the visual comparison templates; The materials without foreign matters screened are transported through each type of belt, the vibration signals of each belt at the corresponding feeding port at each collection time point are collected through the distributed fiber auscultation sensor, the average vibration signals of each belt at the corresponding feeding port at each collection time point are obtained through mean value calculation, the average vibration signals include the average vibration frequency of the roller rotation and the average vibration frequency of the belt, the roller rotation frequency threshold range and the belt vibration frequency range of each belt are set as the average vibration frequency of the roller rotation plus or minus 30% and the average vibration frequency of the belt plus or minus 30% respectively, and the surface temperature and the roller temperature of each belt at each collection time point during the transportation test are recorded through the temperature sensor and the thermal imaging camera, the average surface temperature and the average roller temperature of each belt during the transportation test are obtained through mean value calculation; Step two, double-dimension fusion monitoring: when the material shielding causes visual failure, conveying foreign matter monitoring and tearing monitoring are performed on the belt through the double-dimension monitoring mode of fiber auscultation vibration monitoring and temperature monitoring, and the state corresponding to the belt is comprehensively evaluated; The specific process of performing conveying foreign matter monitoring and tearing monitoring on the belt through fiber auscultation vibration monitoring is as follows: The double-dimension monitoring mode is vibration signal monitoring and temperature value monitoring. When the vision is invalid due to material shielding, the distributed optical fiber auscultation sensor starts to operate first, according to the double-dimension monitoring non-key foreign matter monitoring position corresponding to the belt section where the sensor is located, the sensor collects the rotating frequency of the roller at the feeding port of the belt section where the sensor is located and the vibration frequency of the belt, and compares the rotating frequency of the roller and the vibration frequency of the belt with the rotating frequency threshold of the roller and the vibration frequency recorded in the previous transportation of the material without foreign matter. If high-frequency pulse vibration is detected, and the rotating frequency of the roller and the vibration frequency of the belt are not within the rotating frequency threshold range of the roller and the vibration frequency range, it is determined that the shielded rod-shaped object is determined. If low-frequency continuous vibration occurs, the rotating frequency of the roller and the vibration frequency of the belt are not within the rotating frequency threshold range of the roller and the vibration frequency range, and the amplitude is increased by more than a set percentage compared with the average signal, and the frequency spectrum presents a multi-frequency superposition feature, it is determined that the shielded net-shaped object is determined. If low-frequency strong vibration is captured, the rotating frequency of the roller and the vibration frequency of the belt are not within the rotating frequency threshold range of the roller and the vibration frequency range, the amplitude is increased by more than a set percentage compared with the average signal, and the duration exceeds a set time length, it is determined that the shielded tear is determined. Then, the transportation foreign matter monitoring and tear monitoring are performed through the temperature value monitoring mode. Step three, state correlation evaluation: according to the double-dimension monitoring cooperative identification result, taking the optical fiber auscultation vibration data as the core, the evolution risk of foreign matter to tear is warned within a set time period, and the cause of the belt tear is evaluated. Step four, multi-modal fusion: integrate the visual, optical fiber auscultation vibration and temperature monitoring report, and then form a closed-loop operation and maintenance process.

2. The method for monitoring the condition of a cross-sea stockpile line transport belt according to claim 1, characterized in that, The AI vision recognition method is used to monitor the transportation foreign matter of each belt, and the specific process is as follows: During the belt material transportation process, the industrial cameras and dust-penetrating cameras arranged above the head and tail of each belt and the material falling port collect the belt surface images at a preset collection time interval, start the light supplement device, and through the self-cleaning air device, the AI vision recognition algorithm first calls the visual comparison template in the database, compares the real-time collected image with the reference template, if there is an abnormality after the comparison of the image at a certain collection time point with the reference template image, the image at the collection time point is recorded as a suspected foreign matter image, and the position is recorded as a double-dimension monitoring non-key foreign matter monitoring position, and the position is labeled on the cloud platform and pushed to the mobile APP to remind the operation and maintenance personnel to perform manual monitoring. If there is no abnormality after the comparison of the image at a certain collection time point with the reference template image, it is determined that the actual position of the image at the collection time point is shielded by material, leading to vision failure, and the actual position of the image at the collection time point is recorded as a double-dimension monitoring key foreign matter monitoring position, and then the double-dimension monitoring method is used to monitor the foreign matter at the double-dimension monitoring key foreign matter monitoring position.

3. The method for monitoring the condition of a cross-sea stockpile line transport belt according to claim 2, characterized in that, The AI vision recognition method is used to monitor the tear of each belt, and the specific process is as follows: When the tearing monitoring is performed through the AI visual recognition, the industrial camera photographs the edges and surfaces of each belt in real time according to the preset collection time interval, the AI algorithm calls the edge profile and surface flatness data in the state monitoring reference feature in the non-tearing state, compares the edges and surfaces of each belt photographed in real time with the edge profile and surface flatness data in the state monitoring reference feature in the non-tearing state at the pixel level, detects whether burrs deviating from the reference profile by more than the set burr length threshold value appear, if the detection result shows that burrs deviating from the reference profile by more than the set burr length threshold value appear, the actual position corresponding to the image deviating from the reference profile is recorded as the double-dimension monitoring non-key tearing monitoring position, the position is labeled on the cloud platform, and is pushed to the mobile APP to remind the operation and maintenance personnel to perform manual monitoring, if the detection result does not show that burrs deviating from the reference profile by more than the set burr length threshold value appear, it is determined that material blocking causes visual failure, and the actual position corresponding to the image not deviating from the reference profile is recorded as the double-dimension monitoring key tearing monitoring position, and then the double-dimension monitoring key tearing monitoring position is monitored through the double-dimension monitoring mode.

4. The method for monitoring the condition of a cross-sea stockpile line transport belt according to claim 1, characterized in that, The belt is monitored for transported foreign matter and tearing through the temperature monitoring mode, and the specific process is as follows: The temperature value monitoring mode synchronously works through the thermal imaging camera installed on the roller device bridge under the belt, records the temperature of the blocked area in real time according to the preset collection time interval, compares the temperature with the average belt surface temperature and the average roller temperature obtained through the transportation test, when the double-dimension monitoring key foreign matter monitoring position is monitored, the temperature at the corresponding position changes synchronously with the vibration pulse, presents instantaneous rise and fall, and the temperature difference fluctuation meets the preset pulse temperature difference threshold value based on the average belt surface temperature and the average roller temperature, that is, matches the vibration characteristics of the rod-shaped object, it is determined that the double-dimension monitoring key tearing monitoring position exists the rod-shaped object blocked by the material, if the temperature presents a ladder type increase, the cumulative temperature difference in a unit of time is within the preset ladder type temperature difference range with reference to the reference temperature, and the degree of overall temperature deviation from the average belt surface temperature and the average roller temperature is consistent with the friction characteristics of the mesh-shaped object, it is determined that the double-dimension monitoring key tearing monitoring position exists the mesh-shaped object blocked by the material; If the temperature presents a linear rise when the double-dimension monitoring key tearing monitoring position is monitored, the high-temperature area forms a continuous strip with the belt running and has no pulse fluctuation, and the temperature difference change amplitude meets the preset tearing feature temperature difference range based on the reference temperature, it is determined that the double-dimension monitoring key tearing monitoring position exists the tearing blocked by the material.

5. The method for monitoring the condition of a cross-sea stockpile line transport belt according to claim 4, characterized in that, The state corresponding to the belt is comprehensively evaluated, and the specific process is as follows: If the idler roller rotation frequency and the belt vibration frequency are not within the idler roller rotation frequency threshold range and the belt vibration frequency range, and the difference between the corresponding temperature change and the average temperature meets the temperature matching relationship, the abnormal type is confirmed, and then the specific area of the abnormal position of the fiber stethoscopic sensor on the belt is located. If the idler roller rotation frequency and the belt vibration frequency are not within the idler roller rotation frequency threshold range and the belt vibration frequency range, but there is no corresponding difference between the temperature and the average temperature, it is determined that the material is unevenly disturbed, and no alarm is triggered. If the temperature deviates from the average temperature, but the vibration signal has no corresponding characteristics, it is determined that the material itself has temperature differences, and no alarm is triggered.

6. The method of condition monitoring of a cross-sea stockpile line transport belt according to claim 5, characterized in that, The specific process of the above-mentioned early warning of the evolution risk of the foreign matter to tearing within a set time period is as follows: When early warning of the evolution risk of the foreign matter to tearing within a set time period, by tracking the changes of the vibration and temperature characteristics of the identified foreign matter, and comparing with the state monitoring benchmark characteristics of the test, the vibration soundprint signal of the foreign matter captured by the fiber stethoscopic sensor is monitored in real time, it is determined whether the frequency deviation further expands from the idler roller rotation frequency threshold range and the belt vibration frequency range, and whether the amplitude continuously rises. At the same time, it is observed whether the temperature rise rate changes from pulse type and step type to linear growth. If all the above-mentioned characteristics appear within the set time period, and the low-frequency strong vibration trend characteristic of tearing appears in the vibration spectrum, it is determined that the foreign matter has entered the evolution stage from blocking to friction aggravation to imminent tearing, and an upgraded early warning is triggered immediately, prompting that the belt will be torn.

7. The method for monitoring the condition of a cross-sea stockpile line transport belt according to claim 6, characterized in that, The specific process of the above-mentioned evaluation of the cause of the belt tearing is as follows: According to the two-dimensional monitoring data at the time of tearing, including vibration signal data and temperature value data, and then backtracking the historical records of the tearing set time period, if there is vibration signal meeting the foreign matter characteristics in the historical records of the tearing set time period, and the temperature anomaly at the corresponding position matches the foreign matter characteristics, it is determined that the tearing is caused by foreign matter blocking. If there is no vibration and temperature anomaly related to foreign matter in the historical records of the tearing set time period, only the vibration and temperature characteristics characteristic of tearing exist, it is determined that the tearing is caused by non-foreign matter factors.

8. The method for monitoring the condition of a cross-sea stockpile line based on the state of the transport belt according to claim 7, characterized in that, The specific process of the above-mentioned integration of visual, fiber stethoscopic vibration and temperature monitoring reports is as follows: According to the benchmark characteristics and real-time recognition results of visual monitoring, the abnormal spectrum and frequency deviation data of fiber stethoscopic vibration monitoring, the abnormal temperature rise trend and benchmark temperature difference information of temperature monitoring, combined with the conclusions of foreign matter evolution risk early warning and tearing cause evaluation, the single mode false alarm is eliminated through cross verification, and a comprehensive report containing abnormal type, accurate position, evolution stage and cause is formed. Based on this, the disposal scheme of the operation and maintenance link is promoted, the effect tracking and the benchmark data dynamic updating are tracked, and the closed-loop process from monitoring to solving is completed.

Images