A digital operation and maintenance system and method of a conveying belt machine

Abstract

This invention relates to a digital operation and maintenance system and method for conveyor belts, comprising: laying temperature-sensing optical fibers along both sides of the conveyor belt; connecting the temperature-sensing optical fibers to a temperature-measuring optical fiber host to form a distributed temperature-measuring optical fiber subsystem; also laying fault-auscultating optical fibers along both sides of the belt, connecting them to enhanced-sensitivity inductors arranged at certain intervals; connecting the fault-auscultating optical fibers to a positioning-type auscultating optical fiber host to form a distributed positioning auscultating optical fiber subsystem; and positioning the temperature-sensing and fault-auscultating optical fibers below the left and right idlers. This invention utilizes a distributed detection system to continuously monitor the conveyor belt 24 hours a day, recording the position and status of each monitoring node on the belt, and leveraging the characteristics of distributed monitoring to quickly determine the location of hidden or visible faults.

Description

Technical Field

[0001] This invention relates to a digital operation and maintenance system and method for conveyor belt conveyors, a safety operation monitoring system and method for transportation machinery, and an operation and maintenance system and method for long-distance belt conveyors. Background Technology

[0002] Currently, belt conveyors are mainly used in production lines in industries such as power, mining, metallurgy, and chemicals, playing a crucial role in the production process. If a belt malfunctions, abnormal friction can occur between the belt rollers, idlers, and materials. Prolonged abnormal friction can lead to increased equipment temperature. Furthermore, as dust accumulates between the belt and the idlers, it covers the idler surfaces, worsening heat dissipation and further increasing temperature. Traditionally, conveyor belt maintenance relies on regular manual inspections and maintenance when necessary. Manual inspections are inefficient and cannot provide 24 / 7 monitoring. Conveyor belts are typically installed outdoors at heights of several meters, in confined spaces, posing safety hazards for manual passage, and are susceptible to adverse weather conditions such as strong winds, heavy snow, and high temperatures. During belt operation, issues such as slippage, overheating, full-load restarts after emergency stops, handling flammable and flammable materials, and control circuit safety require inspection and testing. Therefore, how to safely and efficiently monitor the operation of conveyor belts is a problem that needs to be solved. Summary of the Invention

[0003] To overcome the problems of existing technologies, this invention proposes a digital operation and maintenance system and method for conveyor belts. The system and method, tailored to different areas, utilize a power monitoring unit, an AI video analysis unit, a temperature-measuring optical cable, and a fault-detecting optical cable to monitor belt operation data in real time. Combined with an operation and maintenance data model, a belt operation health analysis report is generated. This addresses uncontrollable factors beyond the eight basic faults of traditional conveyor belts, such as pull rope misalignment, smoke, longitudinal tearing, stalling, slippage, and stalling. Furthermore, by providing real-time health status analysis reports on the core auxiliary equipment of the belt—such as the reducer, belt motor, brake disc brake, fan, and oil pump—a more intuitive overall "health check" of the belt is achieved.

[0004] The objective of this invention is achieved as follows: A digital operation and maintenance system for a conveyor belt includes: laying temperature-sensing optical fibers along both sides of the conveyor belt, the temperature-sensing optical fibers being connected to a temperature-measuring optical fiber host to form a distributed temperature-measuring optical fiber subsystem; also laying fault-auscultating optical fibers along both sides of the belt, connecting to enhanced-sensitivity inductors arranged at certain intervals, the fault-auscultating optical fibers being connected to a positioning-type auscultating optical fiber host to form a distributed positioning auscultating optical fiber subsystem; the temperature-sensing optical fibers and fault-auscultating optical fibers being positioned below the left and right idlers; and also laying communication optical cables along the belt, the communication optical cables being connected to key points along the belt. The explosion-proof binocular camera is equipped with infrared thermal imaging and visible light imaging. The communication optical cable also connects the explosion-proof binocular camera and sound monitoring instrument installed around the drum, reducer, motor, and brake disc brake. Each motor, lighting facility, and ventilation facility is equipped with a power monitoring unit. The distributed temperature measurement optical fiber subsystem, the distributed positioning and auscultation optical fiber subsystem, each explosion-proof binocular camera, sound monitoring instrument, and power monitoring unit are connected to the back-end platform equipped with an intelligent temperature analysis unit, an intelligent voiceprint analysis and recognition unit, an intelligent video image recognition unit, an intelligent infrared image analysis unit, an intelligent power analysis unit, and an AI intelligent comprehensive analysis subsystem via optical cables.

[0005] Furthermore, the power monitoring unit includes: a current monitoring sensor, a voltage monitoring sensor, a temperature monitoring sensor, and a humidity monitoring sensor.

[0006] Furthermore, the temperature-sensing optical fiber and the fault auscultation optical fiber are fixed to the conveyor belt frame by steel clips or iron wires.

[0007] Furthermore, the aforementioned backend platform comprises a platform layer, an application layer, a network layer, and a sensor layer:

[0008] The platform layer includes: an operation and maintenance platform, a display and control platform, and a model analysis platform. The model analysis platform includes several components: environmental monitoring, behavior analysis, process control, equipment monitoring, and material monitoring.

[0009] The application layer includes: unified monitoring (comprehensive data aggregation); unified management (overall resource management); unified intelligence (multi-modal AI analysis); unified early warning (situation assessment and early warning); unified operation and maintenance (efficient maintenance and loss prevention); and unified brain (health model analysis).

[0010] Network layer: 4G / 5G network; video network; security monitoring network; Internet of Things (IoT) components;

[0011] The sensor layer consists of several parts: temperature-measuring optical fiber, power acquisition unit, and auscultation optical fiber.

[0012] A digital operation and maintenance method for a conveyor belt conveyor using the above system, comprising the following steps:

[0013] Step 1, collect normal operating parameters: collect the parameters output by each sensor when the belt is running under no-load, under rated load, and within the allowable overload range. The parameters include: temperature changes along the belt obtained by the temperature sensing fiber, acoustic parameters along the belt obtained by the fault auscultation fiber, visible light images, infrared images, power supply voltage and current, temperature rise of each motor, and ambient humidity. Transmit the collected data to the back-end platform via TCP / IP.

[0014] The distributed temperature measurement fiber optic subsystem is used to monitor the temperature changes along the belt in real time. The intelligent temperature analysis unit analyzes the temperature data collected by the temperature sensing fiber, identifies the location and real-time changes of the temperature measurement point, and determines whether it is normal or abnormal. If an abnormality occurs, it analyzes the degree of abnormal temperature change and takes measures according to the degree of change.

[0015] The distributed positioning and auscultation fiber optic subsystem is used to monitor the sound along the belt. The intelligent voiceprint analysis and recognition unit analyzes the voiceprint signals collected by each sensitive inductor head of the belt to identify whether the sound is abnormal.

[0016] The intelligent video image recognition unit identifies and analyzes visible light images to determine the operating status of the belt. Through the images, it can identify whether the belt is started or stopped, whether there are foreign objects on the belt, whether there is material blockage, whether the belt is broken, whether there are personnel illegally climbing over or staying, whether the belt is running off track, large pieces of material on the belt, and material flow rate on the belt.

[0017] Intelligent infrared image analysis unit: Analyzes the heat generated during the operation of the belt conveyor, including: the heat generated by friction between the belt and the idler, the heat generated by the rotation of the idler, the heat generated by friction between the material and the belt, and the heat of the material itself; by analyzing these heats in conjunction with the distributed temperature data obtained by the distributed temperature measurement fiber optic subsystem and the temperature status of various electrical devices, the overall temperature status of the belt is comprehensively analyzed.

[0018] All data, including raw data and analysis data, are aggregated and stored in the database of the backend platform.

[0019] Step 2, Intelligent Parameter Analysis: Through the AI ​​intelligent comprehensive analysis subsystem, the system integrates temperature parameters along the conveyor belt, acoustic parameters along the conveyor belt, visible light images of key points, infrared images of key points, as well as voltage, current, temperature rise of each motor, and ambient humidity parameters to analyze and identify various normal and abnormal operating states of the conveyor belt. This includes: normal and abnormal start-stop states of the equipment; normal and abnormal temperatures along the conveyor belt; normal and abnormal operating sounds; normal and abnormal voltage, current, temperature, and humidity of various electrical equipment; and abnormal events occurring during conveyor belt operation, including: longitudinal tearing, material blockage, foreign objects, belt misalignment, and personnel climbing over guardrails. These normal and abnormal states are input as parameters into the AI ​​intelligent comprehensive analysis subsystem to build an operation and maintenance data model, train the system's ability to identify normal operation and abnormal events, and store the identification results in the database.

[0020] Step 3, Operation Monitoring: During normal operation of the belt, the temperature and sound along the belt are monitored in real time. At the same time, the key nodes of the belt are monitored using visible light and thermal imaging. The voltage, current, temperature and humidity of various electrical facilities are also monitored for safety. The operating status is analyzed in real time, and the current data is continuously compared with the data in the database. If there are any anomalies, similar cases are identified, the differences between the two are compared, and corresponding measures are taken. In addition, a personalized maintenance plan for the belt is specified according to the belt type and operating conditions to ensure the belt is in good operating condition. The belt operation health analysis report is generated by combining the operation and maintenance data model.

[0021] Step 4, Warning: If the temperature along the belt gradually increases, accompanied by abnormal sounds or abnormal electrical facilities, a curve of temperature increase and noise increase or abnormal electrical facilities will be plotted. The AI ​​intelligent comprehensive analysis subsystem will search for similar cases in the database, observe the curve changes, predict the increasing trend of temperature, noise and electrical facilities, and thus predict the failure of the belt conveyor. The increasing curve will be displayed on the display platform as a warning.

[0022] Step 5, Alarm: Pre-set temperature threshold, noise threshold, and normal operating threshold for electrical facilities along the belt; if the threshold is exceeded during belt operation, or if foreign objects or people enter the belt, an audible and visual alarm will be displayed on the platform.

[0023] Step 6, take emergency measures: If the emergency alarm fails to work, stop the vehicle immediately. If there is a fire, start the spraying immediately and begin the fire extinguishing procedure.

[0024] Furthermore, the operation and maintenance data model of the method includes:

[0025] Data services, model development, and deployment services;

[0026] The data services include data management and datasets; data management includes data processing, which includes proofreading, cleaning, enhancement, and annotation; the datasets include infrared, video, image, and big data.

[0027] Model development includes: algorithm development, model training, and model management;

[0028] The algorithm development includes: deep learning, online IED, and remote terminal;

[0029] The model training includes: training cluster, training model, and visualization analysis;

[0030] Model management includes: automated machine learning frameworks, model refinement, model optimization, and model repositories;

[0031] Deployment services include: deployment inference;

[0032] Deployment inference includes side-end inference and online inference.

[0033] The advantages and beneficial effects of this invention are as follows: This invention utilizes a distributed detection system to continuously monitor and supervise belt conveyors 24 / 7, recording the position and status of each monitoring node. Furthermore, it leverages the characteristics of distributed monitoring to quickly pinpoint the location of hidden or overt faults. This invention not only performs individual AI intelligent analysis on the various information acquired by the sensors but also integrates the analyzed data through comprehensive AI intelligent analysis, thereby achieving a comprehensive health diagnosis of the belt and completely solving the problem of predicting faults that cannot be addressed by the existing eight protection controls for belt conveyors. Attached Figure Description

[0034] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0035] Figure 1 This is a schematic diagram of the system described in Embodiment 1 of the present invention;

[0036] Figure 2 This is a schematic diagram of the system described in Embodiment 1 of the present invention. Figure 1 AA section diagram;

[0037] Figure 3 This is a schematic diagram of the structure of the back-end platform of the conveyor belt conveyor as described in Embodiment 4 of the present invention;

[0038] Figure 4 This is a flowchart of the digital operation and maintenance method for conveyor belts as described in Embodiment 5 of the present invention;

[0039] Figure 5 This is a schematic diagram of the visualization development model structure of the digital operation and maintenance method for conveyor belts described in Embodiment Six of the present invention. Detailed Implementation

[0040] Example 1:

[0041] This embodiment is a digital operation and maintenance system for conveyor belts, such as... Figure 1 , 2 As shown. This embodiment includes: laying temperature-sensing optical fibers 2 on both sides of the conveyor belt 1, the temperature-sensing optical fibers being connected to a temperature-measuring optical fiber host to form a distributed temperature-measuring optical fiber subsystem; also laying fault-auscultating optical fibers 3 on both sides of the belt, connecting to enhanced-sensitivity inductors 4 arranged at certain intervals, the fault-auscultating optical fibers being connected to a positioning-type auscultating optical fiber host to form a distributed positioning auscultating optical fiber subsystem; the temperature-sensing optical fibers and fault-auscultating optical fibers are positioned below the left and right idlers 101 and 102; and also laying communication optical cables 5 along the belt, the communication optical cables being connected to infrared thermal imaging devices installed at key points along the belt. The explosion-proof binocular camera 6, which uses visible light imaging, is connected by a communication optical cable to the explosion-proof binocular camera and sound monitoring device installed around the drum, reducer, motor, and brake disc brake. Each motor, lighting facility, and ventilation facility is equipped with a power monitoring unit. The distributed temperature measurement optical fiber subsystem, the distributed positioning and auscultation optical fiber subsystem, each explosion-proof binocular camera, sound monitoring device, and power monitoring unit are connected to the back-end platform equipped with an intelligent temperature analysis unit, an intelligent voiceprint analysis and recognition unit, an intelligent video image recognition unit, an intelligent infrared image analysis unit, an intelligent power analysis unit, and an AI intelligent comprehensive analysis subsystem via optical cables.

[0042] This embodiment addresses various aspects of belt conveyors by employing temperature-measuring fiber optics, fault-auscultation fiber optics, power monitoring units, and explosion-proof binocular cameras to monitor belt operation data in real time. Combined with an operational data model, it generates a belt operation health analysis report, resolving uncontrollable factors beyond the traditional eight major protective rope faults such as belt misalignment, smoke, longitudinal tearing, stalling, slippage, and stalling. Furthermore, it provides a more intuitive overall "health check" of the belt by analyzing the real-time health status of core auxiliary equipment such as the reducer, belt motor, brake disc brake, fan, and oil pump.

[0043] Applying different monitoring technologies in different scenarios can enable digital operation and maintenance of belts and provide technical solutions, including: power consumption monitoring and analysis of belt control systems and auxiliary monitoring systems, visible light and thermal imaging video analysis and monitoring of the visible area above the belt, and the comprehensive application of temperature measurement optical cable and sound diagnostic optical cable technology in the non-visual area below the belt.

[0044] The digital operation and maintenance system for conveyor belts is an important measure for achieving intelligent and refined management of conveyor belts. It plays a crucial role in improving the availability of conveyor belt equipment and ensuring safe production. It also has significant value in reducing operation and maintenance costs and meeting environmental protection requirements. The following are the core technologies of the digital operation and maintenance system for conveyor belts:

[0045] 1) Real-time temperature monitoring: The surface temperature of key components such as belts, rollers, idlers, drive motors, and bearings is monitored 24 hours a day using infrared thermal imagers and temperature-measuring optical fibers to prevent excessive temperature from damaging the belts and causing fire risks.

[0046] 2) Remote monitoring and data integration: Enables data acquisition, equipment control, measurement and parameter adjustment, and real-time data monitoring visualization.

[0047] 3) Belt condition monitoring: Image recognition technology is used to detect foreign objects that may be above the belt and belt misalignment. The power monitoring unit is used to monitor the status of the belt motor and auxiliary equipment in real time.

[0048] 4) Intelligent early warning and maintenance management: The system uses data analysis and deep learning algorithms to predict equipment failures, issue early warning signals in advance, and guide maintenance personnel to carry out preventive maintenance, thereby reducing equipment downtime and maintenance costs.

[0049] This embodiment mainly includes: real-time temperature monitoring, real-time sound monitoring, power monitoring unit module, explosion-proof binocular camera configuration, and monitoring of various major auxiliary facilities. It also includes intelligent analysis at the back-end operation and maintenance end, intelligent analysis of distributed temperature measurement, and intelligent analysis of distributed location monitoring, ultimately realizing the digital operation and maintenance functions of the transmission belt conveyor.

[0050] Power monitoring unit:

[0051] It is the data acquisition and control unit for digital operation and maintenance of belt conveyors. It transmits the acquired data to the back-end platform via TCP / IP. The belt conveyor includes the acquisition of various signals such as current, voltage, temperature, and humidity from the motor, control system, monitoring equipment, and supplementary lighting equipment.

[0052] Explosion-proof binocular camera:

[0053] The binocular camera is primarily used for conveyor belt temperature measurement and hotspot tracking. High-definition visible light, combined with backend AI algorithms, enables intelligent detection under various conveyor belt conditions. Different focal lengths adapt to different monitoring distances. The camera uses 5MP high-definition visible light, providing realistic and detailed images that capture different details of the conveyor belt during operation. It supports a temperature measurement range of -20℃ to 550℃. Visible light detail and thermal imaging measurement are perfectly integrated. The camera incorporates multiple algorithms for filtering and identifying people and vehicles, and features robust perimeter protection. It supports abnormal temperature alarms, fire monitoring algorithms, and various temperature measurement rules. Infrared + white light dual illumination provides uniform lighting and penetrates darkness. The binocular camera supports audible and visual alarms, remote voice intercom, fire and lightning protection, and PoE power supply, playing a core monitoring role in the digital operation and maintenance of conveyor belts.

[0054] The belt conveyor dual-lens camera system can be deployed in 4, 6, or 8 units, with thermal imagers installed at various locations. The thermal imagers enable alarm linkage and support remote voice communication, adapting to real-time monitoring of the belt's operating status.

[0055] Back-end intelligent analysis device:

[0056] Through AI-powered intelligent video analytics, abnormal situations such as conveyor belt start / stop status, abnormal temperature, longitudinal tearing and blockage, foreign objects, belt misalignment, and personnel climbing over guardrails can be identified. Real-time safety monitoring and early warning of key nodes can be provided, improving inspection efficiency, reducing the risk of accidents, and assisting coal mining enterprises in their digital and intelligent transformation and upgrading.

[0057] Distributed temperature measurement fiber optic subsystem:

[0058] The equipment consists of a temperature-sensing optical fiber, a temperature-measuring optical fiber host, and platform software. The Distributed Temperature Measurement Fiber System (DTS) utilizes the principles of Raman scattering (RAMAN) and optical time-domain reflectometry (OTOR) during light transmission in optical fibers to acquire spatial temperature distribution information. When the external temperature changes, the temperature-sensing optical fiber transmits the temperature change signal to the host for intelligent analysis, triggering an alarm for abnormal temperature changes and performing secondary video verification through platform linkage. Distributed temperature sensors have the following advantages:

[0059] Long distance and wide coverage: Continuous real-time monitoring of temperature over tens of kilometers, enabling distributed measurement.

[0060] High precision and accurate measurement: Temperature resolution of 0.1℃, sub-meter level positioning accuracy, and support for measuring temperature information at multiple points.

[0061] Fully passive, conformal installation: The front-end passive device design is not constrained by power supply conditions, focusing on temperature measurement inside confined spaces and long-distance temperature measurement of linear objects.

[0062] High reliability and light-sensitive material: unaffected by external electromagnetic interference, inherently safe, and posing no risk of combustion or explosion. Suitable for humid, high and low temperature environments, and resistant to ammonia, chlorine, salt spray, acid rain, etc.

[0063] Distributed temperature-sensing optical fibers offer advantages such as self-testing capability, positioning accuracy within ±1m, high accuracy, short response time, low construction difficulty, and low project cost. The temperature-sensing optical cable is installed below the conveyor belt rollers, and the cable wiring is laid inside the I-beam beside the conveyor belt, without affecting the conveyor belt's operation.

[0064] Distributed positioning and auscultation fiber optic subsystem:

[0065] This device consists of an enhanced fiber optic cable, a positioning-type auscultation fiber optic host, and platform software. All sound is generated by vibration, and all vibrations are formed and propagated as sound waves. Therefore, fiber optics can detect vibration sources and reconstruct sound waves, enabling remote auscultation of abnormal sounds from conveyor belt rollers. This eliminates the need for on-site personnel to inspect conveyor belts in poor condition, improving maintenance efficiency. Its main advantages include:

[0066] Adaptable to various locations and miniature sensing: The distributed sensing passive design is highly adaptable to the environment and resistant to interference.

[0067] Sound restoration and intelligent demodulation: Built-in high-performance DSP module for multimodal voiceprint recognition with low error and false alarm.

[0068] Enhanced sensitivity design, meter-level accuracy: unique Technology optimizes signal-to-noise ratio, enhances sensitivity for industrial diagnostic detection, and enables high-precision measurement.

[0069] AI algorithms and scene adaptation: a continuously enriched sample library, online self-learning, and rapid iteration for scene optimization.

[0070] Fiber optic cables are installed in contact with the steel belt frame, and sensitive inductors are laid every 5-10m. The sensitive sensors are usually laid in locations with stable connections, rigid structures, and good conductivity.

[0071] The installation of a distributed positioning and auscultation system completely solves the problem of delayed shutdown caused by the eight protection delays during the operation of belt conveyors.

[0072] The installation of a distributed positioning and auscultation system completely solves the problem of delayed shutdown caused by the eight protection delays during the operation of belt conveyors.

[0073] Example 2:

[0074] This embodiment is an improvement upon Embodiment 1, detailing the power monitoring unit. The power monitoring unit in this embodiment includes: a current monitoring sensor, a voltage monitoring sensor, a temperature monitoring sensor, and a humidity monitoring sensor.

[0075] The personnel reduction monitoring unit monitors the electrical facilities of the belt conveyor, mainly detecting whether the power consumption exceeds the load. The temperature sensor mainly detects the temperature of the belt drive motor, its corresponding motor control circuit, and the hydraulic system (mainly the braking device) of the belt conveyor, so as to form an electrical operating curve, thereby predicting and judging the operating status of the belt conveyor.

[0076] Example 3:

[0077] This embodiment is an improvement upon the above embodiment, detailing the installation method of the temperature-sensing optical fiber and the fault-auscultation optical fiber. In this embodiment, the temperature-sensing optical fiber and the fault-auscultation optical fiber are fixed to the conveyor belt frame using steel clips and wire.

[0078] Temperature-sensing optical fibers are very thin and need to be secured with taut wires and steel clips to prevent them from falling off or even being torn apart.

[0079] Example 4:

[0080] This embodiment is an improvement upon the above embodiments, detailing the backend platform. The backend platform described in this embodiment is installed in an industrial control computer or other electronic digital processing device with digital storage and computing capabilities. The backend platform includes a platform layer, an application layer, a network layer, and a sensor layer, such as... Figure 3 As shown.

[0081] The platform layer includes: an operation and maintenance platform, a display and control platform, and a model analysis platform. The model analysis platform includes several parts: environmental monitoring, behavior analysis process control equipment monitoring, and material monitoring.

[0082] The platform layer is the basic operating architecture installed on the computer operating system. The operation and maintenance platform is the basic operating layer, the display and control platform consists of peripherals such as monitors, keyboards, and mice, and the model analysis platform is an AI intelligent processing device used to intelligently analyze and judge the data collected by all sensors.

[0083] The application layer includes: unified monitoring (comprehensive data aggregation); unified management (overall resource management); unified intelligence (multi-modal AI analysis); unified early warning (situation assessment and early warning); unified operation and maintenance (efficient maintenance and loss prevention); and unified brain (health model analysis).

[0084] The application layer is an intelligent interactive architecture used for various monitoring and maintenance of devices.

[0085] Network layer: 4G / 5G network; video network; security monitoring network; Internet of Things (IoT) components.

[0086] The network layer serves as a channel for communication within the belt conveyor and with external devices. It can communicate via an internal network or a wireless network to connect with external equipment.

[0087] The sensor layer consists of several parts: temperature-measuring optical fiber, power acquisition unit, and auscultation optical fiber.

[0088] The sensor layer is mainly used for the configuration and connection of various sensors, as well as the preprocessing of each sensor, including AI intelligent analysis of the data acquired by each sensor.

[0089] Example 5:

[0090] This embodiment describes a digital operation and maintenance method for a conveyor belt conveyor using the aforementioned system. The steps of the method are as follows, and the process is as follows: Figure 4 As shown:

[0091] Step 1: Collect normal operating parameters: Collect the parameters output by each sensor when the belt is running under no-load, under rated load, and within the allowable overload range. Parameter items include: temperature change along the belt obtained by the temperature sensing fiber, acoustic parameters along the belt obtained by the fault auscultation fiber, visible light image, infrared image, power supply voltage and current, temperature rise of each motor, and ambient humidity. Transmit the collected data to the back-end platform via TCP / IP.

[0092] The distributed temperature measurement fiber optic subsystem is used to monitor temperature changes along the belt in real time. The intelligent temperature analysis unit analyzes the temperature data collected by the temperature sensing fiber, identifies the location and real-time changes of the temperature measurement point, and determines whether it is normal or abnormal. If an abnormality occurs, it analyzes the degree of abnormal temperature change and takes measures according to the degree of change, including: pre-alarm, alarm, and emergency alarm.

[0093] The distributed positioning and auscultation fiber optic subsystem is used to monitor the sound along the belt. The intelligent voiceprint analysis and recognition unit analyzes the voiceprint signals collected by each sensitive inductor head of the belt to identify whether the sound is abnormal. The sound categories include: the sound of the roller rubbing against the belt, the sound of the belt running off track, the sound of the roller bearing, and the sound of the belt rubbing against the material.

[0094] Explosion-proof binocular cameras acquire visible light and infrared images.

[0095] The intelligent video image recognition unit identifies and analyzes visible light images to determine the operating status of the belt. Through the images, it can identify whether the belt is started or stopped, whether there are foreign objects on the belt, whether there is material blockage, whether the belt is broken, whether there are personnel illegally climbing over or staying, whether the belt is running off track, large pieces of material on the belt, and the material flow rate on the belt.

[0096] Intelligent infrared image analysis unit: Analyzes the heat generated during the operation of the belt conveyor, including: the heat generated by friction between the belt and the idler, the heat generated by the rotation of the idler, the heat generated by friction between the material and the belt, and the heat of the material itself; by analyzing these heats in conjunction with the distributed temperature data obtained by the distributed temperature measurement fiber optic subsystem and the temperature status of various electrical devices, a comprehensive analysis of the overall temperature status of the belt operation is performed.

[0097] Intelligent power analysis unit: It judges the overall status of the belt conveyor by measuring the overall voltage and current, the voltage and current of each electrical facility, the ambient temperature, and the temperature rise of the facility itself.

[0098] All data, including raw and analyzed data, is aggregated and stored in the database of the backend platform.

[0099] Step 2, Intelligent Parameter Analysis: Through the AI ​​intelligent comprehensive analysis subsystem, the system integrates temperature parameters along the conveyor belt, acoustic parameters along the conveyor belt, visible light images of key points, infrared images of key points, as well as voltage, current, temperature rise of each motor, and ambient humidity parameters to analyze and identify various normal and abnormal operating states of the conveyor belt. This includes: normal and abnormal start-stop states of the equipment; normal and abnormal temperatures along the conveyor belt; normal and abnormal operating sounds; normal and abnormal voltage, current, temperature, and humidity of various electrical equipment; and abnormal events occurring during conveyor belt operation, including: longitudinal tearing, material blockage, foreign objects, belt misalignment, and personnel climbing over guardrails. These normal and abnormal states are input as parameters into the AI ​​intelligent comprehensive analysis subsystem to build an operation and maintenance data model, train the system's ability to identify normal operation and abnormal events, and store the identification results in the database.

[0100] Step 3, Operation Monitoring: During normal operation of the belt, the temperature and sound along the belt are monitored in real time. At the same time, key nodes of the belt are monitored using visible light and thermal imaging. Voltage, current, temperature, and humidity of various electrical facilities are also monitored for safety. The operating status is analyzed in real time, and the current data is continuously compared with the data in the database. If any anomalies are found, similar cases are identified, the differences between the two are compared, and corresponding measures are taken. As needed, a belt health analysis report is generated, and a personalized maintenance plan for the belt is specified according to the belt type and operating conditions to ensure good belt operation. The belt operation health analysis report is generated in conjunction with the operation and maintenance data model.

[0101] This paper addresses several key issues: First, the difficulty in identifying the start / stop status of belt conveyors during operation. Second, single-function video cameras cannot provide real-time assessment of belt operation and material blockage. Third, relying on blockage sensors for emergency stopping after material accumulation at the conveyor outlet is too delayed, by which time material buildup has already occurred. Fourth, abnormal belt temperatures are difficult to detect visually, and excessively high belt speeds can lead to abnormal temperatures and potential fires. Fifth, longitudinal tears in the belt can easily cause personal injury. Sixth, belt misalignment and pull rope issues only appear after on-site personnel alarm, and current pull rope and misalignment sensors cannot provide timely feedback on the belt's operation, making them outdated. This paper utilizes digital belt conveyor maintenance technology, employing distributed temperature and sound monitoring to monitor the overall condition of the belt, including temperature and noise change curves. AI continuously compares this data with historical data to determine if the belt is operating normally and whether potential faults are possible, thus resolving the problem of untimely detection of pull rope and misalignment issues.

[0102] An important method for operational monitoring is the use of belt operation reports and maintenance plans. Thanks to the distributed real-time monitoring system, real-time data on belt operation status can be quickly acquired, and the system can predict potential changes in belt operation status through operating curves, thereby determining a suitable maintenance plan for the belt conveyor.

[0103] Step 4, Warning: If the temperature along the belt gradually increases, accompanied by abnormal sounds or abnormal electrical facilities, a curve of temperature increase and noise increase or abnormal electrical facilities will be plotted. The AI ​​intelligent comprehensive analysis subsystem will search for similar cases in the database, observe the curve changes, predict the increasing trend of temperature, noise and electrical facilities, and thus predict the failure of the belt conveyor. The increasing curve will be displayed on the display platform as a warning.

[0104] Early warning involves detecting potential problems in belt conveyors, such as a problem with a bearing on an idler roller that doesn't affect the normal operation of the belt but causes increased bearing noise and temperature. The increase in noise and temperature is gradual, not instantaneous. By recording time on the horizontal axis and noise and temperature on the vertical axis, the pattern of the increase can be identified, thus predicting the period until the bearing can function normally again.

[0105] The key to this step lies in the crucial role of the AI ​​intelligent comprehensive analysis subsystem. The judgment of the current event is formed by comparing it with past events, rather than simply providing feedback on the event. Instead, it involves AI analysis, confirmation and processing of similar past events, and the determination of current countermeasures.

[0106] Step 5, Alarm: Pre-set the temperature threshold, noise threshold, and normal operating threshold of electrical facilities along the belt; if the threshold is exceeded during belt operation, or if foreign objects or people enter the belt, an audible and visual alarm will be displayed on the platform.

[0107] Alarms are usually triggered by emergencies, such as foreign objects appearing on the belt or someone illegally climbing over the belt. The reaction of these phenomena to the sensors is a sudden increase in the temperature and noise of the belt at the point of the incident. Distributed sensors can quickly locate the location of the incident.

[0108] Step 6, take emergency measures: If the emergency alarm fails to work, stop the vehicle immediately. If there is a fire, start the spraying immediately and begin the fire extinguishing procedure.

[0109] After an emergency occurs, the system will quickly sound an alarm. If there is no response after the alarm has been running for a period of time, the system will automatically take emergency measures, such as shutting down the machine and turning off all power except for the backup emergency power supply. If there is a fire, the system will activate the fire suppression spraying program.

[0110] Example 6:

[0111] This embodiment is an improvement on Embodiment 5, and is a refinement of the operation and maintenance data model of the operation and maintenance method in Embodiment 5.

[0112] Algorithm model development services are tailored to specific client needs, creating visualized development models based on actual business data, such as... Figure 5 As shown.

[0113] 1) Real-time temperature monitoring: Using an infrared thermal imager and temperature measuring fiber optic cable, the surface temperature of key parts such as belts, rollers, drive units, and bearings is measured 24 hours a day to prevent safety accidents such as belt damage and fire caused by excessive temperature.

[0114] 2) Belt condition monitoring: Image recognition technology is used to identify foreign objects that may be present on the belt and to monitor belt misalignment in real time. The power monitoring unit is used to monitor the status of the belt motor and auxiliary equipment in real time.

[0115] 3) Intelligent early warning and maintenance management: The system uses data analysis and deep learning algorithms to predict equipment failures, issue early warning signals in advance, and guide maintenance personnel to carry out preventive maintenance, thereby reducing equipment downtime and maintenance costs.

[0116] 4) Remote monitoring and data acquisition: Enables data acquisition, equipment control, measurement, parameter adjustment, and real-time data control visualization management.

[0117] The visualization development model of the method described in this embodiment includes:

[0118] Data services, model development, and deployment services.

[0119] The data services include data management and datasets. Data management includes data processing, which includes proofreading, cleaning, enhancement, and annotation. The datasets include infrared data, video data, image data, and big data data.

[0120] In addition to the raw data collected, there is a large amount of data that has been analyzed and judged, as well as past data. Therefore, storage and management of the data, as well as the processed datasets, are required so that AI can perform analysis and comparison. Therefore, data services are very important.

[0121] Model development includes: algorithm development, model training, and model management.

[0122] The algorithm development includes: deep learning, online IED, and remote terminal;

[0123] The model training includes: training cluster, training model, and visualization analysis;

[0124] Model management includes: automated machine learning frameworks, model refinement, model optimization, and model repositories;

[0125] Deployment services include: deployment inference;

[0126] Deployment inference includes side-end inference and online inference.

[0127] Algorithm Development:

[0128] Deep learning employs neural network structures, such as convolutional neural networks (CNN) and recurrent neural networks (RNN).

[0129] Neural networks: related to deep learning, used to process complex data streams and nonlinear relationships.

[0130] Basic machine learning (ML) algorithms include regression, classification, and clustering.

[0131] Model training:

[0132] Training cluster: Jointly train the model on multiple nodes to improve the model's accuracy and robustness.

[0133] Training the model: Using data to fit and optimize the model so that it can better predict the target variable.

[0134] Visualization analysis: Use charts and visualization tools to help analyze the performance and effectiveness of the model during training.

[0135] Model Management:

[0136] Automated machine learning framework: used to automatically select the optimal algorithm and parameters, reducing human intervention.

[0137] Model refinement: Monitor and optimize the model to ensure its continued performance.

[0138] Model optimization: Improve model performance by adjusting hyperparameters or feature engineering.

[0139] Model repository: Stores all trained models for later use.

[0140] Deployment service:

[0141] Deploying inference: Deploying the model to specific devices for inference operations, such as inference on a server or real-time processing on IoT devices.

[0142] Deploying inference includes side-end inference and online inference. Side-end inference may refer to performing inference on local computing resources, while online inference refers to performing inference over a network in a real-time environment (such as a network environment).

[0143] Models play a central role in machine learning and artificial intelligence. This is achieved through the following steps:

[0144] Data preparation:

[0145] Collect high-quality training data.

[0146] Clean and preprocess the data in advance to remove noise.

[0147] Algorithm Design:

[0148] Choose a suitable algorithm (such as CNN, RNN).

[0149] Design the network architecture to ensure that the desired features are captured.

[0150] Model training:

[0151] Use the training data to fit the model and optimize the model parameters.

[0152] The training process and results are evaluated through visual analysis.

[0153] Model Management:

[0154] Automatically select the optimal algorithm and hyperparameters (automatic machine learning).

[0155] Regularly monitor model performance to ensure continuous optimization.

[0156] Deployment reasoning:

[0157] Deploy the model into the application for inference.

[0158] Build side-end inference and online inference functions to achieve real-time processing.

[0159] Model Application:

[0160] Enables data processing of images, text, or time series.

[0161] Apply models to solve practical problems, such as identifying targets or predicting trends.

Claims

1. A digital operation and maintenance system for conveyor belts, characterized in that, include: Temperature-sensing optical fibers are laid along both sides of the conveyor belt. These fibers are connected to a temperature-measuring optical fiber host, forming a distributed temperature-measuring optical fiber subsystem. Fault-sensing optical fibers are also laid along both sides of the belt, connected to enhanced-sensitivity inductors spaced at regular intervals. These fault-sensing optical fibers are connected to a positioning-type auscultation optical fiber host, forming a distributed positioning auscultation optical fiber subsystem. The temperature-sensing and fault-sensing optical fibers are positioned below the left and right idlers. Communication optical cables are also laid along the belt, connecting to explosion-proof structures equipped with infrared thermal imaging and visible light cameras installed at key points along the belt. The binocular camera, along with the communication optical cable, is connected to an explosion-proof binocular camera and a sound monitoring device installed around the drum, reducer, motor, and brake disc brake. Each motor, lighting facility, and ventilation facility is equipped with a power monitoring unit. The distributed temperature measurement fiber optic subsystem, the distributed positioning and auscultation fiber optic subsystem, each explosion-proof binocular camera, the sound monitoring device, and the power monitoring unit are connected via optical cables to a back-end platform equipped with an intelligent temperature analysis unit, an intelligent voiceprint analysis and recognition unit, an intelligent video image recognition unit, an intelligent infrared image analysis unit, an intelligent power analysis unit, and an AI intelligent comprehensive analysis subsystem.

2. The system according to claim 1, characterized in that, The power monitoring unit includes: a current monitoring sensor, a voltage monitoring sensor, a temperature monitoring sensor, and a humidity monitoring sensor.

3. The system according to claim 2, characterized in that, The aforementioned temperature-sensing optical fiber and fault-auscultation optical fiber are fixed to the conveyor belt frame by steel clips or iron wires.

4. The system according to claim 2, characterized in that, The aforementioned backend platform comprises a platform layer, an application layer, a network layer, and a sensor layer. The platform layer includes: an operation and maintenance platform, a display and control platform, and a model analysis platform. The model analysis platform includes several components: environmental monitoring, behavior analysis, process control, equipment monitoring, and material monitoring. The application layer includes: unified monitoring (comprehensive data aggregation); unified management (overall resource management); unified intelligence (multi-modal AI analysis); unified early warning (situation assessment and early warning); and unified operation and maintenance (efficient maintenance and loss prevention). Unified Brain: A Health Model Analysis of Several Components; Network layer: 4G / 5G network; The components include: video network; security monitoring network; and the Internet of Things. The sensor layer consists of several parts: temperature-measuring optical fiber, power acquisition unit, and auscultation optical fiber.

5. A digital operation and maintenance method for a conveyor belt conveyor using the system described in claim 4, characterized in that, The steps of the method are as follows: Step 1, collect normal operating parameters: collect the parameters output by each sensor when the belt is running under no-load, under rated load, and within the allowable overload range. The parameters include: temperature changes along the belt obtained by the temperature sensing fiber, acoustic parameters along the belt obtained by the fault auscultation fiber, visible light images, infrared images, power supply voltage and current, temperature rise of each motor, and ambient humidity. Transmit the collected data to the back-end platform via TCP / IP. The distributed temperature measurement fiber optic subsystem is used to monitor the temperature changes along the belt in real time. The intelligent temperature analysis unit analyzes the temperature data collected by the temperature sensing fiber, identifies the location and real-time changes of the temperature measurement point, and determines whether it is normal or abnormal. If an abnormality occurs, it analyzes the degree of abnormal temperature change and takes measures according to the degree of change. The distributed positioning and auscultation fiber optic subsystem is used to monitor the sound along the belt. The intelligent voiceprint analysis and recognition unit analyzes the voiceprint signals collected by each sensitive inductor head of the belt to identify whether the sound is abnormal. The intelligent video image recognition unit identifies and analyzes visible light images to determine the operating status of the belt. Through the images, it can identify whether the belt is started or stopped, whether there are foreign objects on the belt, whether there is material blockage, whether the belt is broken, whether there are personnel illegally climbing over or staying, whether the belt is running off track, large pieces of material on the belt, and material flow rate on the belt. Intelligent infrared image analysis unit: Analyzes the heat generated during the operation of the belt conveyor, including: the heat generated by friction between the belt and the idler, the heat generated by the rotation of the idler, the heat generated by friction between the material and the belt, and the heat of the material itself; by analyzing these heats in conjunction with the distributed temperature data obtained by the distributed temperature measurement fiber optic subsystem and the temperature status of various electrical devices, the overall temperature status of the belt is comprehensively analyzed. All data, including raw data and analysis data, are aggregated and stored in the database of the backend platform. Step 2, Intelligent Parameter Analysis: Through the AI ​​intelligent comprehensive analysis subsystem, the system integrates temperature parameters along the conveyor belt, acoustic parameters along the conveyor belt, visible light images of key points, infrared images of key points, as well as voltage, current, temperature rise of each motor, and ambient humidity parameters to analyze and identify various normal and abnormal operating states of the conveyor belt. This includes: normal and abnormal start-stop states of the equipment; normal and abnormal temperatures along the conveyor belt; normal and abnormal operating sounds; normal and abnormal voltage, current, temperature, and humidity of various electrical equipment; and abnormal events occurring during conveyor belt operation, including: longitudinal tearing, material blockage, foreign objects, belt misalignment, and personnel climbing over guardrails. These normal and abnormal states are input as parameters into the AI ​​intelligent comprehensive analysis subsystem to build an operation and maintenance data model, train the system's ability to identify normal operation and abnormal events, and store the identification results in the database. Step 3, Operation Monitoring: During normal operation of the belt, the temperature and sound along the belt are monitored in real time. At the same time, the key nodes of the belt are monitored using visible light and thermal imaging. The voltage, current, temperature and humidity of various electrical facilities are also monitored for safety. The operating status is analyzed in real time, and the current data is continuously compared with the data in the database. If there are any anomalies, similar cases are identified, the differences between the two are compared, and corresponding measures are taken. In addition, a personalized maintenance plan for the belt is specified according to the belt type and operating conditions to ensure the belt is in good operating condition. The belt operation health analysis report is generated by combining the operation and maintenance data model. Step 4, Warning: If the temperature along the belt gradually increases, accompanied by abnormal sounds or abnormal electrical facilities, a curve of temperature increase and noise increase or abnormal electrical facilities will be plotted. The AI ​​intelligent comprehensive analysis subsystem will search for similar cases in the database, observe the curve changes, predict the increasing trend of temperature, noise and electrical facilities, and thus predict the failure of the belt conveyor. The increasing curve will be displayed on the display platform as a warning. Step 5, Alarm: Pre-set temperature threshold, noise threshold, and normal operating threshold for electrical facilities along the belt; if the threshold is exceeded during belt operation, or if foreign objects or people enter the belt, an audible and visual alarm will be displayed on the platform. Step 6, take emergency measures: If the emergency alarm fails to work, stop the vehicle immediately. If there is a fire, start the spraying immediately and begin the fire extinguishing procedure.

6. The method according to claim 5, characterized in that, The operation and maintenance data model of the method includes: Data services, model development, and deployment services; The data services include data management and datasets; data management includes data processing, which includes proofreading, cleaning, enhancement, and annotation; the datasets include infrared, video, image, and big data. Model development includes: algorithm development, model training, and model management; The algorithm development includes: deep learning, online IED, and remote terminal; The model training includes: training cluster, training model, and visualization analysis; Model management includes: automated machine learning frameworks, model refinement, model optimization, and model repositories; Deployment services include: deployment inference; Deployment inference includes side-end inference and online inference.

Images