A method and device for early warning and dynamic control of blockage prevention in underground coal mine belt conveyors

Abstract

This invention discloses a method and device for early warning and dynamic control of blockage prevention in underground coal mine belt conveyors, belonging to the field of mine safety and automation control technology. It involves deploying a multi-source sensor array to synchronously collect real-time load, operating speed, and material accumulation image data, and performing joint analysis based on dynamic matching degree calculation and material morphology recognition. Based on the analysis results, a multi-level early warning logic generates an early warning signal and triggers the early warning device. Depending on the early warning level, local adaptive speed adjustment or remote collaborative control is executed. Local control adjusts the belt speed based on the dynamic matching degree, while remote control receives and executes commands from the ground platform via fiber optic communication. This invention achieves early warning and intelligent dynamic control of blockage risks, improving the safety and operational efficiency of underground transportation systems.

Description

Technical Field

[0001] This invention relates to the field of mine safety and automation control technology, and in particular to a method, device, equipment and storage medium for anti-blockage early warning and dynamic control of underground belt conveyors in coal mines. Background Technology

[0002] In underground coal mine material transportation systems, belt conveyors are core equipment, and their operational stability directly affects mine production efficiency. Current belt conveyor blockage monitoring technologies used in the industry generally suffer from the following technical bottlenecks: 1. Insufficient monitoring accuracy: Existing systems mostly use a single type of sensor (such as infrared ranging sensor or ultrasonic level sensor) to determine blockage. These sensors are easily affected by high concentrations of coal dust underground, humidity fluctuations (relative humidity often reaches 60%-95%), and differences in material particle size, resulting in distorted monitoring signals and a false alarm rate of over 30%. They cannot accurately distinguish between "normal material accumulation" and "blockage risk". 2. Rigid control logic: The system lacks an adaptive control strategy that dynamically matches the real-time transport volume. The belt speed is mostly fixed (usually set at 1.0-1.5 m / s). When the material transport volume increases sharply (such as during high-production periods in coal mining), the fixed speed cannot disperse the material in time, easily leading to blockages at the conveyor head or chute due to overload. Conversely, when the transport volume decreases sharply (such as during face maintenance), fixed-speed operation results in wasted motor energy, with an energy redundancy of 15%-25%. 3. Delayed response mechanism: When the traditional system detects the risk of material blockage, it can only provide on-site audible and visual alarms. It requires the inspection personnel to arrive on-site and manually stop or adjust the machine. The average response time from alarm to handling exceeds 15 minutes, during which the blockage area is prone to expand, and even secondary failures such as belt misalignment and motor burnout may occur. 4. Lack of Remote Control: The existing system lacks a complete remote data transmission and control link. The ground dispatch center cannot obtain real-time belt operation parameters (such as real-time load, speed, and sensor status), and can only grasp the situation through reports from underground personnel. Troubleshooting requires multiple trips between the underground and the surface, resulting in low efficiency and an average troubleshooting time exceeding 2 hours. Therefore, there is an urgent need to develop a blockage prevention and early warning system with multi-source sensing fusion, dynamic adaptive control, and remote collaborative handling capabilities to overcome the above-mentioned technical bottlenecks. Summary of the Invention

[0003] The present invention aims to at least partially solve one of the technical problems in the related art.

[0004] To address this issue, this invention discloses a method for early warning and dynamic control of blockage prevention in underground coal mine belt conveyors. It utilizes multi-source sensors to synchronously collect real-time load, speed, and material image data, calculates the dynamic matching degree, and performs joint analysis using material morphology recognition. Based on the analysis results, a multi-level early warning logic generates an early warning signal and triggers an early warning device. Depending on the warning level, local adaptive speed adjustment or remote collaborative control is executed. Local control adjusts the belt speed, while remote control receives and executes commands from the ground platform. This method achieves early warning and dynamic control of blockage risks, improving the safety and efficiency of the transportation system.

[0005] Another objective of this invention is to provide a device for anti-blockage early warning and dynamic control of underground belt conveyors in coal mines.

[0006] The third objective of this invention is to provide a computer device.

[0007] The fourth objective of this invention is to provide a non-transitory computer-readable storage medium.

[0008] To achieve the above objectives, this invention proposes a method for anti-blockage early warning and dynamic control of underground belt conveyors in coal mines, comprising: S1, deploy a multi-source sensor array and synchronously collect real-time load data, running speed data and material accumulation image data of the belt conveyor; S2 calculates the dynamic matching degree of load and speed based on real-time load data and running speed data, and performs joint analysis by combining the material accumulation pattern recognition results output by the image recognition algorithm. S3, based on the joint analysis results of dynamic matching degree and material accumulation pattern recognition, generates warning signals of corresponding warning levels through multi-level warning logic and triggers corresponding warning devices. S4 executes local adaptive speed control and remote collaborative handling control according to the warning level; among them, local adaptive speed control adjusts the belt running speed according to the dynamic matching degree, and remote collaborative handling control receives and executes the handling instructions from the ground platform through the fiber optic communication link.

[0009] The anti-blockage early warning and dynamic control method for underground belt conveyors in coal mines according to an embodiment of the present invention may also have the following additional technical features: In one embodiment of the present invention, the deployment of a multi-source sensor array and the synchronous acquisition of real-time load data, operating speed data, and material accumulation image data of the belt conveyor include: S11, Install the resistance strain gauge weight sensor on the idler roller bracket 10m upstream of the unloading port of the belt conveyor head and every 50m in the middle section, and set the sampling frequency to 1 time / second; S12, install the explosion-proof high-definition camera for mining at 50cm above the head chute and at the top of the roadway in the middle section where material is prone to blockage. The lens is facing the conveyor belt material conveyor surface, and the image sampling frequency is set to 5 frames / second.

[0010] In one embodiment of the present invention, calculating the dynamic matching degree of load and speed based on real-time load data and operating speed data includes: S21, using formula Calculate the dynamic matching degree; where, As the reference load, As the reference speed; S22, performs grayscale conversion on the image data using the formula. The grayscale value is calculated, and Gaussian filtering is used to remove coal dust interference noise; where R, G, and B are the pixel values ​​of the three channels of the image.

[0011] In one embodiment of the present invention, a warning signal is generated based on the joint analysis results of dynamic matching degree and material accumulation pattern recognition results, including: S31, when the dynamic matching degree And image recognition results , and When the alarm is triggered, a level one warning is activated and the sound and light alarm is controlled to emit a sound signal at a frequency of 800Hz, while the yellow warning light flashes at a frequency of 1 time per second. S32, when the dynamic matching degree and When the alarm is triggered, a level-two warning is activated and the audible and visual alarm is activated to emit a sound signal at a frequency of 1000Hz. At the same time, the orange warning light flashes at a frequency of 2 times per second, and the warning information is pushed to the ground platform.

[0012] In one embodiment of the present invention, performing local adaptive speed control based on the warning level includes: S41, when the dynamic matching degree At that time, a speed adjustment command is sent to the frequency converter to reduce the belt speed from the rated speed. Down to ; S42, when the first or second level warning is triggered, the belt speed is reduced from... Down to It gradually recovers to its original value at a gradient of 0.1 m / s. .

[0013] In one embodiment of the present invention, it further includes: S5 monitors the communication status and data validity of each sensor in real time through the sensor fault self-diagnosis module. If no sensor data is received for 5 consecutive seconds, it is determined to be a communication fault, automatically switches to the backup sensor, and sends sensor fault status information to the ground platform.

[0014] To achieve the above objectives, another aspect of the present invention provides a blockage prevention early warning and dynamic control device for underground belt conveyors in coal mines, comprising: The multi-source sensor deployment and data synchronization acquisition module is used to deploy a multi-source sensor array and synchronously acquire real-time load data, running speed data, and material accumulation image data of the belt conveyor. The dynamic matching degree calculation and image joint analysis module is used to calculate the dynamic matching degree of load and speed based on real-time load data and running speed data, and to perform joint analysis by combining the material accumulation pattern recognition results output by the image recognition algorithm. The multi-level early warning signal generation and triggering module is used to generate early warning signals of corresponding early warning levels and trigger corresponding early warning devices based on the joint analysis results of dynamic matching degree and material accumulation pattern recognition results through multi-level early warning logic. The local speed regulation and remote collaborative control module is used to perform local adaptive speed regulation control and remote collaborative handling control according to the warning level. The local adaptive speed regulation control adjusts the belt running speed according to the dynamic matching degree, and the remote collaborative handling control receives and executes the handling instructions from the ground platform through the fiber optic communication link.

[0015] In one embodiment of the present invention, it further includes: The sensor fault self-diagnosis module is used to monitor the communication status and data validity of each sensor in real time. If no sensor data is received for 5 consecutive seconds, it is determined to be a communication fault, automatically switches to the backup sensor, and sends sensor fault status information to the ground platform.

[0016] This invention discloses a method and device for anti-blockage early warning and dynamic control of underground belt conveyors in coal mines, which effectively overcomes the shortcomings of existing technologies, such as reliance on a single signal, delayed early warning, and rigid control. Through multi-source data fusion and joint analysis, it achieves early and accurate identification of blockage risks and multi-level early warning. The adaptive speed regulation and remote collaborative control mechanism based on early warning levels significantly improves the system's response speed and handling flexibility, strengthens the proactive safety control capabilities of the underground transportation process, ensures continuous and stable operation, and reduces the intensity of manual intervention and the risk of equipment blockage, thus possessing significant engineering application value.

[0017] To achieve the above objectives, a third aspect of this application provides a computer device, including a processor and a memory; wherein the processor reads executable program code stored in the memory to run a program corresponding to the executable program code, for implementing a method for anti-blocking early warning and dynamic control of a coal mine underground belt conveyor as described in the first aspect embodiment.

[0018] To achieve the above objectives, the fourth aspect of this application proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a method for anti-blockage early warning and dynamic control of a coal mine underground belt conveyor as described in the first aspect embodiment.

[0019] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0020] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of a method for anti-blocking early warning and dynamic control of underground belt conveyors in coal mines according to an embodiment of the present invention; Figure 2 This is an overall architecture diagram of a coal mine underground belt conveyor anti-blockage early warning and dynamic control system according to an embodiment of the present invention; Figure 3 This is a control flowchart of a method for anti-blocking early warning and dynamic control of underground belt conveyors in coal mines according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of an anti-blockage early warning and dynamic control device for an underground belt conveyor in a coal mine according to an embodiment of the present invention; Figure 5 It is a computer device according to an embodiment of the present invention. Detailed Implementation

[0021] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0022] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0023] The following description, with reference to the accompanying drawings, describes a method, apparatus, equipment, and storage medium for anti-blockage early warning and dynamic control of underground belt conveyors in coal mines, according to embodiments of the present invention.

[0024] The core idea of ​​this invention is to construct a multimodal monitoring system that integrates mechanical status and visual perception by deploying a multi-source sensor array to synchronously collect real-time load, operating speed, and material accumulation image data. Based on the dynamic matching degree calculation of load and speed, and combined with the material accumulation pattern recognition results output by the image recognition algorithm, a joint analysis is performed to achieve accurate and early quantitative assessment of the belt conveyor's operating status and the risk of material blockage. Based on this joint analysis result, the system automatically generates and triggers corresponding level warning signals through multi-level warning logic, and then executes local adaptive speed adjustment or remote collaborative handling control according to the warning level. This solution transforms traditional passive response and experience-based handling relying on manual intervention into an intelligent proactive prevention and control closed loop based on multi-source information fusion, real-time quantitative risk assessment, graded warning triggering, and dynamic execution of control strategies. Ultimately, it significantly improves the safety and reliability of the underground transportation system while ensuring the continuity and efficiency of production operations.

[0025] Example 1 To achieve the above invention, embodiments of the present invention provide a method for anti-blockage early warning and dynamic control of underground belt conveyors in coal mines, such as... Figure 1 As shown, it includes: S1 deploys a multi-source sensor array and synchronously collects real-time load data, operating speed data, and material accumulation image data of the belt conveyor.

[0026] Specifically, this step establishes the basic capabilities of the system's perception layer through the collaborative deployment of multi-source heterogeneous sensors and a data synchronization acquisition mechanism, providing high-quality, real-time data input for subsequent load speed matching calculations and AI image recognition.

[0027] Specifically, this step employs three types of sensors for coordinated deployment: (1) a resistance strain gauge weight sensor (model: YZC133) with a range of 0~500 kg / m and an accuracy class of 0.1%FS. These sensors are installed 10m upstream of the discharge port at the head of the belt conveyor and on the idler roller supports every 50m in the middle section, with a total of 35 sensors deployed (the number can be adjusted according to the belt length). These sensors are used to collect the material load value per unit length of the belt in real time. (Unit: kg / m); (2) Hall effect speed sensor (model: HGC1030), with a measurement range of 0~2 m / s and a resolution of 0.01 m / s, is installed on the end of the tail roller shaft and the belt running speed is calculated by detecting the roller speed. (Unit: m / s); (3) Mine explosion-proof high-definition camera (model: KBA127), resolution 1920×1080, frame rate 25fps, dustproof and waterproof rating IP68, installed 50cm above the machine head chute and the top of the roadway in the middle section where material is easily blocked, with the lens facing the conveyor belt material conveying surface, used to collect material accumulation images.

[0028] Furthermore, the sampling frequency of the weight and speed sensors is 1 time / second, and the sampling frequency of the image sensor is 5 frames / second. All sensor data is transmitted to the local control unit (LCU) via an RS485 bus at a transmission rate of 115200bps, with a data transmission delay ≤100 ms. In addition, image data requires preprocessing after acquisition, including grayscale conversion (formula: ) and Gaussian filtering (kernel size) , This is to remove noise interference from coal dust and ensure the accuracy of image recognition.

[0029] Specifically, this step is applicable to belt conveyor systems in complex underground coal mine environments, especially under conditions of high coal dust concentration, fluctuating humidity (60%-95%), and significant differences in material particle size. By fusing multi-source data, it effectively enhances the system's ability to perceive the material accumulation state. The collected data will be used for load-speed matching. The calculation (where) , It combines an improved YOLOv5 image recognition model to assess material blockage risk, thereby achieving graded early warning and adaptive speed control.

[0030] Specifically, through the coordinated deployment and synchronous acquisition of multiple sensors, the system can acquire real-time load, speed, and image information of the belt conveyor, providing reliable data support for subsequent intelligent analysis and control decisions. Compared with traditional single-sensor monitoring methods, this solution significantly improves monitoring accuracy and environmental adaptability, laying a solid foundation for early identification and rapid response to material blockage risks.

[0031] Furthermore, S1 includes: S11. Install the resistance strain gauge weight sensor on the idler roller bracket 10m upstream of the discharge port of the belt conveyor head and every 50m in the middle section, and set the sampling frequency to 1 time / second.

[0032] Specifically, this sensor employs the strain gauge principle, measuring the minute deformation of the belt under material load, and then converting this deformation into the weight of material per unit length (denoted as ). (Unit: kg / m). Its measuring range is 0–500 kg / m, and its accuracy class is 0.1%FS, meeting the measurement needs of high-load and high-dust environments in underground coal mines.

[0033] Furthermore, the sensors need to be fixed to the load-bearing structure of the idler roller bracket, ensuring they are perpendicular to the belt's running trajectory to obtain accurate load distribution data. The installation spacing is designed to be one sensor every 50m in the middle section, combined with a sensor 10m upstream of the discharge port at the machine head, forming a load monitoring network covering the entire transport path. This arrangement effectively captures the dynamic distribution characteristics of materials during transport, providing high spatiotemporal resolution data support for subsequent load-speed matching calculations.

[0034] Furthermore, the sampling frequency is set to 1 time / second, i.e. This frequency ensures real-time data transmission while accommodating system processing power and data transmission bandwidth limitations. All sensor data is transmitted to the local control unit (LCU) via RS485 bus at a rate of 115200 bps, with transmission delay controlled within 100ms, ensuring data synchronization and timely system response.

[0035] Specifically, this step plays a core role in the perception layer of the system, providing a basis for subsequent load speed matching. Provide key input parameters and ,in , Using this as a baseline value, the system can accurately identify the material accumulation state through the fusion analysis of multi-point load data, providing a reliable basis for graded early warning and adaptive control, thereby significantly improving the accuracy and response efficiency of anti-blockage early warning.

[0036] S12, install the explosion-proof high-definition camera for mining at 50cm above the head chute and at the top of the roadway in the middle section where material is prone to blockage. The lens is facing the conveyor belt material conveyor surface, and the image sampling frequency is set to 5 frames / second.

[0037] Specifically, the technical implementation of this step is based on an adaptive design to the underground environment and the real-time requirements of image acquisition. The camera's installation position must strictly adhere to the geometric relationship between the roadway structure and the belt conveyor's trajectory. An installation height of 50cm above the machine head chute ensures the lens can clearly capture the material accumulation near the belt unloading port, avoiding image distortion or obstruction due to angular deviations. In the middle section, where material blockage is prone to occur, the camera is installed at the top of the roadway, with the lens facing the belt conveyor surface to cover the dynamic distribution of material during transportation. Vibration-resistant brackets and fixing bolts must be used during installation to ensure stable operation of the equipment in the underground vibration environment, while meeting the requirements of the explosion-proof rating Exd I Mb, complying with the installation specifications for explosion-proof equipment in the "Coal Mine Safety Regulations" (GB 3836.1-2010).

[0038] Furthermore, the sampling frequency of the image sensor is set to 5 frames per second, which is an optimized configuration based on the system's response speed requirements to changes in material accumulation. A sampling frequency of 5 frames per second can effectively reduce the computational load of data transmission and processing while ensuring image continuity, and ensuring synchronization with the 1 frame per second sampling frequency of the weight and velocity sensors in the time dimension, facilitating subsequent multi-source data fusion analysis.

[0039] Specifically, it primarily serves as the input data acquisition module for the image recognition algorithm (YOLOv5). Through real-time acquisition of high-definition images, the system can identify material accumulation areas and conveyor belt edges, and calculate the bulk density. and matching degree with load speed The system jointly assesses the risk level of material blockage. Its technological value lies in providing the system with intuitive and reliable visual data, compensating for the limitations of a single sensor in complex environments, thereby improving the accuracy and response efficiency of material blockage identification.

[0040] S2 calculates the dynamic matching degree of load and speed based on real-time load data and running speed data, and performs joint analysis by combining the material accumulation pattern recognition results output by the image recognition algorithm.

[0041] Specifically, this step involves fusing real-time data from weight and velocity sensors to construct a load-velocity matching model, and then performing joint analysis with material accumulation morphology features identified by image recognition algorithms, thereby improving the accuracy of early warnings and response efficiency.

[0042] Specifically, this step first involves collecting the material load value per unit length of the belt conveyor using a resistance strain gauge weight sensor (model: YZC133). (Unit: kg / m), and the belt running speed is obtained by a Hall effect speed sensor (model: HGC1030). (Unit: m / s). The Local Control Unit (LCU) calls the load-speed matching algorithm module, according to the formula... Calculate the dynamic matching degree between load and speed. ,in For rated load, This is the rated speed. This formula reflects the combined operating status of the current load and speed, and is used to determine whether the system is approaching or exceeding its capacity.

[0043] Meanwhile, the image recognition algorithm, based on the YOLOv5 model, performs real-time analysis of material accumulation images captured by a high-definition camera (model: KBA127), identifying three types of targets: "material area," "belt edge," and "accumulation area." Image preprocessing includes grayscale conversion (formula: ) and Gaussian filtering (kernel size) , To eliminate coal dust interference noise. The proportion of accumulation areas in the identification results. As a morphological feature parameter, and the matching degree Used in conjunction with risk assessment.

[0044] Furthermore, the load sensor has an accuracy of 0.1%FS, the velocity sensor has a resolution of 0.01m / s, the image sensor has a resolution of 1920×1080, a frame rate of 25fps, and a sampling frequency of 5 frames / second. The LCU performs validity checks on the sensor data, replacing invalid data with the average value of the previous 5 seconds to ensure data continuity and reliability.

[0045] Specifically, this step applies to the real-time monitoring system of underground belt conveyors in coal mines, especially during high-production periods or maintenance periods at the coal mining face. It can dynamically adjust the control strategy according to changes in material transport volume, avoiding material blockage or energy waste caused by fixed-speed operation.

[0046] Specifically, through load-speed matching degree proportion of accumulation patterns The joint analysis significantly improved the accuracy of material blockage risk identification, reduced the false alarm rate to below 10%, and provided key basis for subsequent graded early warning and adaptive speed control, thereby realizing intelligent and real-time anti-blockage early warning and response.

[0047] Furthermore, S2 includes: S21, using formula Calculate the dynamic matching degree; where, As the reference load, This is the reference speed.

[0048] Specifically, this step is based on real-time collected belt load data. (Unit: kg / m) and operating speed (Unit: m / s), compared with the preset reference load and reference speed By comparing the current operating status with the standard operating conditions, the degree of matching can be quantified.

[0049] Specifically, the local control unit (LCU) receives real-time load data collected by the weight sensor (model: YZC133) via the analog input module (SM1233). The belt running speed is obtained through a Hall effect speed sensor (model: HGC1030). The LCU's built-in load-speed matching algorithm module processes the above data in real time, using the formula... Calculate dynamic matching degree This formula compares the product of the actual load and speed with the product of a reference value to reflect whether the current belt operation is within a reasonable load and speed matching range.

[0050] Furthermore, the reference load Compared with the reference speed It is designed based on the typical operating conditions of underground belt conveyors in coal mines and is representative of the industry. The weight sensor has an accuracy class of 0.1%FS, and the speed sensor has a resolution of 0.01m / s, ensuring... and The acquisition error is controlled within 1%. The LCU's data processing frequency is 1Hz, and the matching degree calculation delay is ≤100ms, meeting the real-time control requirements.

[0051] Specifically, this step is mainly used to determine whether the current belt operation is close to or exceeds the blockage risk threshold. For example, when And image recognition parameters When the system detects an abnormal situation, it will trigger a level-two warning, prompting the ground platform to intervene. Through the calculation of dynamic matching degree, the system can achieve a quantitative assessment of the belt's operating status, providing a basis for subsequent graded warnings and adaptive speed regulation.

[0052] Specifically, by introducing a joint load and speed matching model, the system's sensitivity and accuracy in identifying material blockage risks are effectively improved. Compared to traditional fixed-speed control methods, this solution can dynamically adjust the belt speed according to real-time load, reducing motor energy consumption and improving material dispersal efficiency, thereby achieving the goals of intelligent, energy-saving, and efficient transportation control.

[0053] S22, performs grayscale conversion on the image data using the formula. The grayscale value is calculated, and Gaussian filtering is used to remove coal dust interference noise; where R, G, and B are the pixel values ​​of the three channels of the image.

[0054] Specifically, in the image data processing step, the acquired material conveying image is first subjected to grayscale conversion to reduce image dimensionality and enhance the robustness of subsequent image recognition algorithms. The grayscale conversion uses a weighted average method, and its core formula is: ,in , , These represent the pixel values ​​of the red, green, and blue channels of a pixel in the image, respectively, with values ​​ranging from [0, 255]. This formula is designed based on the differences in human eye sensitivity to different color channels, with the green channel having the highest weight, followed by red, and then blue, thus achieving image grayscale while preserving visual perception information.

[0055] Specifically, the grayscale conversion process is executed in real time by the image processing module in the local control unit (LCU) after the image sensor acquires the original RGB image. This module is typically deployed inside a mine-use explosion-proof PLC and performs pixel-by-pixel calculations on each frame of the image using embedded image processing algorithms to ensure that the image processing latency is controlled within a specified range. Within this range, to meet the needs of real-time monitoring.

[0056] Furthermore, to eliminate image noise caused by underground coal dust, a Gaussian filter is used to smooth the grayscale image. The kernel size of the Gaussian filter is set to... Standard deviation This parameter combination demonstrated good noise suppression capabilities in experimental verification, effectively removing high-frequency noise caused by coal dust particles without significantly blurring image edges. The Gaussian filter's convolution kernel weight distribution follows a two-dimensional Gaussian function, and its calculation method is as follows: ; Specifically, this step mainly processes images of the areas above the head chute and the middle section of the belt conveyor that are prone to material blockage, with an image resolution of [resolution value missing]. The frame rate is The grayscale and filtered images will be used as input to the YOLOv5 model to identify key targets such as material accumulation areas and conveyor belt edges, thereby providing visual evidence for judging the risk of material blockage.

[0057] Specifically, the technical effect of this step is to significantly improve image quality and recognition accuracy, especially in environments with high coal dust concentrations, where it can effectively suppress noise interference and reduce the false recognition rate. Combined with subsequent AI image recognition algorithms, this processing flow provides a high-quality visual data foundation for realizing a "multi-source perception fusion" anti-blocking and early warning system, and is a key link in the system to achieve intelligent and real-time early warning.

[0058] S3, based on the joint analysis results of dynamic matching degree and material accumulation pattern recognition, generates warning signals of corresponding warning levels through multi-level warning logic and triggers corresponding warning devices.

[0059] Specifically, this step is the core link in the system to realize intelligent early warning and rapid response. Its technical implementation is based on the embedded logic judgment module and early warning signal driving circuit in the local control unit (LCU).

[0060] Furthermore, dynamic matching degree It is composed of real-time load (Unit: kg / m) and real-time speed (Unit: m / s) Calculated together, where For rated load, The rated speed. Material accumulation pattern recognition results. This is achieved by using an improved YOLOv5 model to perform target detection and classification on material accumulation images acquired by an image sensor. The ratio of the accumulated area to the belt edge is calculated to characterize the degree of material accumulation. The early warning logic is based on... and The joint status is classified into three levels, specifically including Level 1, Level 2 and Level 3 warnings.

[0061] Furthermore, the early warning logic module running inside the LCU first receives data from the load-speed matching algorithm module. Values ​​and outputs of the AI ​​image recognition module Value. According to the preset multi-level early warning rules, when... and ,or and When the situation is assessed as a Level 1 warning (low risk), the local audible and visual warning device is triggered, with the sound frequency at 800Hz and the light flashing yellow at a frequency of 1 time per second; when and When the situation is assessed as a Level II alert (medium risk), a local audible and visual alert is triggered (sound frequency 1000Hz, orange flashing, frequency 2 times / second), and an alert is simultaneously pushed to the ground remote control platform (GRCP); if the Level II alert is triggered within 10 seconds... and If the risk level remains above the threshold, the alert will be upgraded to Level 3 (high risk), triggering a constant red light and a 1200Hz audible alarm. GRCP will also automatically display a pop-up notification and send an SMS to the administrator.

[0062] Specifically, this step is widely used in underground belt conveyor systems in coal mines, especially in critical locations such as the unloading port at the head of the conveyor and areas prone to blockages in the middle section. The LCU and the ground platform exchange data via a fiber optic communication link, with a communication delay of ≤1 second, ensuring the real-time nature of early warning signals and timely response of remote control.

[0063] Specifically, this step significantly improves the accuracy and response efficiency of early warnings through multi-source data fusion and a tiered early warning mechanism. Compared to traditional systems, this solution reduces the early warning response time to less than 2 seconds, and the remote handling time for level 3 early warnings is ≤3 seconds, effectively reducing the risk of material blockage expansion due to delayed early warnings and improving the intelligence level and operational safety of the coal mine transportation system.

[0064] Furthermore, S3 includes: S31, when the dynamic matching degree And image recognition results , and When the alarm is triggered, a Level 1 warning is activated and the audible and visual alarm is activated to emit an audible signal at a frequency of 800Hz, while the yellow warning light flashes at a frequency of 1 time per second.

[0065] Specifically, this step is one of the core early warning mechanisms based on multi-source sensing data fusion and adaptive control strategies. It aims to provide timely local response when potential material blockage risks are initially identified, preventing the risks from escalating further.

[0066] Furthermore, dynamic matching degree It is calculated through the load-speed matching algorithm module, and the formula is as follows: ,in This represents the material load per unit length of the current belt (collected by a weight sensor). The current belt speed (collected by the speed sensor). For rated load, At rated speed. Image recognition results. The output of the improved YOLOv5 model represents the confidence level of the material accumulation area in the current image. Its value ranges from [0, 1], and the higher the value, the more severe the accumulation.

[0067] Furthermore, the local control unit (LCU) will control the audible and visual alarm to emit an 800Hz sound signal, while the yellow warning light flashes at a frequency of 1 time per second. The audible and visual alarm adopts a mining-grade explosion-proof design, conforming to the GB3836.1-2010 standard "Electrical Apparatus for Explosive Atmospheres," ensuring safe operation in high-dust and high-humidity underground environments. The yellow light flashes at a frequency of 1Hz, meeting the sensitivity requirements of human vision for warning signals, and can attract the attention of on-site personnel in a short period of time.

[0068] Specifically, this step is mainly used for immediate early warning of belt conveyors in low-risk material blockage situations, such as minor material accumulation or a mismatch between load and speed that has not yet reached a serious level. In this case, the system responds quickly with local audible and visual warnings, without waiting for remote commands, thereby achieving millisecond-level initial handling and reducing the risk of the fault spreading.

[0069] Specifically, by setting clear early warning thresholds and combining load speed matching with image recognition results, the system can achieve early identification and graded response to material blockage risks. The triggering conditions for the first-level early warning are designed to take into account the complementarity of load and image information, effectively improving the accuracy and timeliness of the early warning, and providing a basic guarantee for subsequent medium and high-risk handling. This demonstrates the innovation and practicality of this invention in intelligent and real-time early warning.

[0070] S32, when the dynamic matching degree and When the alarm is triggered, a level-two warning is activated and the audible and visual alarm is activated to emit a sound signal at a frequency of 1000Hz. At the same time, the orange warning light flashes at a frequency of 2 times per second, and the warning information is pushed to the ground platform.

[0071] Specifically, this early warning mechanism is based on a comprehensive judgment of load velocity matching algorithm and image recognition results, exhibiting high technical integration and response efficiency. The local control unit (LCU) calculates the load velocity matching degree in real time. ,in This represents the material load per unit length of the current belt (collected by a weight sensor). The current belt speed (collected by the speed sensor). For rated load, This is the rated speed. Simultaneously, the image recognition module, based on an improved YOLOv5 model, identifies the material accumulation area and calculates the current material accumulation ratio. This is the ratio of the area of ​​the accumulation zone to the effective conveying area of ​​the belt. When and At that time, the system determined that the alert status was Level 2.

[0072] Furthermore, after a Level 2 warning is triggered, the audible and visual alarm will... The system emits an audio signal at a frequency selected within the range of human hearing sensitivity to ensure that underground workers can quickly detect it. Simultaneously, the orange warning light... The warning signal flashes at a specific frequency, using high-brightness LEDs (compliant with GB3836.1-2010 mining explosion-proof standard) to provide a visual warning. The warning information is transmitted via a fiber optic communication module. The data is uploaded to the ground remote control platform (GRCP) at a certain rate, and the information push delay is controlled within [the specified range]. This ensures that the ground dispatch center can promptly grasp the situation on site.

[0073] Specifically, this step applies to scenarios where underground coal mine belt conveyors experience localized material accumulation but have not yet formed a complete blockage, especially when the material transport volume is large and the belt speed cannot be adjusted in time. For example, during high-production periods at the coal face, if the belt speed is not reduced in time, material accumulates in the head or middle idler roller area. The system will quickly identify this risk through multi-source data fusion and respond synchronously through local audible and visual alarms and remote information push to prevent further deterioration of the blockage.

[0074] Specifically, this step enables rapid identification and multimodal early warning of medium-risk levels, effectively shortening the time from risk identification to early warning response, and controlling the average response time to within [a certain range]. Within this range. Simultaneously, by pushing early warning information to ground platforms, data support was provided for remote intervention, enhancing the overall intelligence and collaborative handling capabilities of the system and significantly reducing the risk of secondary failures caused by delayed early warnings.

[0075] S4 executes local adaptive speed control and remote collaborative handling control according to the warning level; among them, local adaptive speed control adjusts the belt running speed according to the dynamic matching degree, and remote collaborative handling control receives and executes the handling instructions from the ground platform through the fiber optic communication link.

[0076] Specifically, this step dynamically senses the belt's operating status and material accumulation, and combines this with an early warning level judgment mechanism to achieve adaptive adjustment of the belt speed or execution of remote control commands, thereby effectively reducing the risk of material blockage and improving system operating efficiency and safety.

[0077] Specifically, this step is divided into two control modes: local adaptive speed control and remote collaborative handling control. Local adaptive speed control is based on load-speed matching. The calculation results are used to adjust the speed, among which This represents the current material load per unit length (collected by a weight sensor). The current belt speed (collected by the speed sensor). and The baseline value set for the system. When and ( When the percentage of material accumulation area in the image recognition algorithm is reached, the system determines it to be a high-risk state, and the local control unit (LCU) reduces the belt speed from the rated speed. Down to This improves material evacuation efficiency and reduces the probability of blockage. During recovery, the system employs a linear recovery strategy with a recovery rate of [value missing]. This ensures smooth speed changes and avoids material deviation or secondary blockage.

[0078] Furthermore, remote collaborative response and control interacts with the ground-based remote control platform (GRCP) via a fiber optic communication link. When the system triggers a Level 3 warning ( and (If the delay exceeds 10 seconds), the ground platform will automatically pop up a prompt and provide operation options such as "remote shutdown," "adjust unloading rate," and "manual speed adjustment." After the administrator selects an option, the control command is transmitted to the LCU via a fiber optic link, with an execution delay not exceeding [a certain value]. To ensure rapid response, if "remote shutdown" is selected, the LCU will immediately cut off the power to the belt motor and shut down the unloading device, while simultaneously reporting the shutdown status to the ground platform. The feedback delay will not exceed [a certain value]. .

[0079] Specifically, this procedure is widely applicable to belt conveyor systems in complex underground coal mine environments, especially during high-production periods at the coal face or equipment maintenance. It can automatically adjust operating strategies based on real-time load and image recognition results, reducing manual intervention. Its technical value lies in significantly improving system response speed and control accuracy, shortening fault handling time, reducing the risk of secondary accidents, and enhancing the system's autonomous operation capability in the event of communication interruptions through local and remote dual-mode control, thereby improving overall reliability and scalability.

[0080] Furthermore, S4 includes: S41, when the dynamic matching degree At that time, a speed adjustment command is sent to the frequency converter to reduce the belt speed from the rated speed. Down to .

[0081] Specifically, this step is a key component of the adaptive speed control logic in this invention, aiming to optimize energy consumption and improve operating efficiency under low-load conditions. The local control unit (LCU) is based on real-time acquired weight sensor data. With speed sensor data The current dynamic matching degree is calculated through the load speed matching algorithm module. The formula for calculating the matching degree is as follows: ,in For rated load, This is the rated speed. When When the value is below the set threshold of 0.5, the LCU determines that the current state is low load and the belt speed needs to be reduced to decrease motor energy consumption. At this time, the LCU sends a speed control command to the frequency converter through the digital output module (SM1222) to adjust the speed to... For example, when hour, .

[0082] Furthermore, the key parameters involved in this step include rated speed. Dynamic matching degree Adjusted speed The LCU's control response time is ≤100ms, ensuring real-time speed control commands. The inverter supports a frequency adjustment range of 0.1Hz to 50Hz, corresponding to a speed adjustment range of 0.05m / s to 1.5m / s, meeting the speed reduction requirements in this step. Furthermore, the system supports the Modbus communication protocol, ensuring compatibility and stability with the inverter.

[0083] Specifically, this step applies to the low-load operation phase of underground belt conveyor systems in coal mines, such as during coal face maintenance and when material output decreases. By reducing the belt speed, the system can effectively reduce the motor's no-load running time, lower energy consumption, and prevent material spillage or belt idling caused by excessive speed, thereby improving the system's economic efficiency and safety.

[0084] Specifically, this step enables intelligent speed control of the belt conveyor under low load conditions, significantly reducing motor energy consumption and improving system operating efficiency. Combined with multi-source sensor data fusion and dynamic matching algorithms, the system possesses stronger environmental adaptability and control precision, providing strong support for energy conservation, consumption reduction, and intelligent operation of underground coal mine transportation systems.

[0085] S42, when the first or second level warning is triggered, the belt speed is reduced from... Down to It gradually recovers to its original value at a gradient of 0.1 m / s. .

[0086] Specifically, this step is based on load-speed matching degree The calculation results, combined with the real-time image recognition algorithm, show the recognition rate of material accumulation areas. After comprehensive assessment, if a risk of material blockage is confirmed, the system will automatically reduce the belt speed from the current operating speed. Reduce to the preset emergency speed This speed threshold is the optimal value obtained through extensive experimental verification. It can reduce the material accumulation rate while maintaining the basic continuity of belt conveyor, thereby effectively improving material dispersal efficiency by about 40% and reducing the risk of material blockage by up to 70%.

[0087] Furthermore, the LCU communicates with the inverter's digital output module (SM1222) to send speed adjustment commands, controlling the output frequency of the belt motor to achieve precise speed regulation. The speed regulation process employs a linear recovery strategy, meaning that once the risk of material blockage is confirmed to have been mitigated (…), the speed adjustment continues. and After 5 seconds, the system gradually reduces the velocity from [previous speed] at a gradient of 0.1 m / s. Restore to The setting of this recovery rate takes into account the inertial distribution characteristics of the material on the belt, avoiding material slippage or deviation caused by a sudden increase in speed, thereby ensuring the stability of the transportation process.

[0088] Specifically, the response time for speed regulation must be controlled within 2 seconds to ensure the system reacts quickly to sudden material blockage risks. Simultaneously, the control accuracy of the frequency converter should reach 0.01 m / s to meet the refined requirements of speed regulation. The communication delay between the LCU and the frequency converter should be ≤100 ms, complying with the safety standards for real-time control systems in underground coal mines.

[0089] Specifically, this step is widely used in underground coal mine belt conveyor systems, especially during high-production periods at the coal face or when material particle size varies significantly. Through intelligent speed regulation, the system can effectively cope with instantaneous load fluctuations, reduce material blockage accidents caused by speed mismatch, and improve transportation efficiency and equipment operation safety.

[0090] Specifically, this step enables dynamic response control of material blockage risk, significantly improving the system's intelligence level. Through rapid deceleration and smooth recovery of speed, the system can effectively reduce equipment overload risk, extend the service life of the belt conveyor, and reduce secondary failures caused by material blockage, such as belt misalignment and motor overheating, while ensuring transportation efficiency. It has significant engineering practical value.

[0091] S5 monitors the communication status and data validity of each sensor in real time through the sensor fault self-diagnosis module. If no sensor data is received for 5 consecutive seconds, it is determined to be a communication fault, automatically switches to the backup sensor, and sends sensor fault status information to the ground platform.

[0092] Specifically, this module monitors the communication status and data validity of each sensor in real time, ensuring that the system can still maintain basic monitoring and control capabilities when sensors malfunction, thereby improving the overall system reliability and fault tolerance.

[0093] Specifically, the sensor fault self-diagnosis module, based on the real-time data acquisition and processing capabilities of the PLC (Programmable Logic Controller), continuously polls the communication status of each sensor. Each sensor (including weight, speed, and image sensors) is connected to the local control unit (LCU) via an RS485 bus, with a data transmission rate of 115200bps and communication latency controlled within 100ms. The module sets a data reception time window; if no valid data is received from a sensor within 5 consecutive seconds, it determines that the sensor has experienced a "communication failure." At this time, the system automatically switches to the corresponding backup sensor. The backup sensor is consistent with the primary sensor in hardware model, installation location, and sampling frequency to ensure data continuity and consistency.

[0094] Furthermore, the fault determination threshold is 5 seconds of no data reception, a switching response time of ≤1 second, and a data recovery delay of ≤2 seconds after the backup sensor is switched over. Simultaneously, the module supports the Modbus communication protocol, ensuring compatibility and standardized access with various types of sensors.

[0095] Specifically, this module is widely used in the complex electromagnetic environment and high dust concentration conditions of underground coal mines. Especially when communication is interrupted due to coal dust covering, loose wiring, or power failure, it can respond quickly and maintain system operation. Its linkage mechanism with the ground remote control platform (GRCP) can upload "sensor fault" status information in real time, which makes it easy for the ground dispatch center to grasp the equipment status and arrange maintenance in a timely manner.

[0096] Specifically, the technical benefits of this step lie in significantly improving the system's fault tolerance and operational continuity. Through the automatic switching mechanism, the system can quickly resume data acquisition after a sensor failure, preventing the entire monitoring system from failing due to a single point of failure. Simultaneously, real-time uploading of fault information helps shorten troubleshooting time, improves operational efficiency, and reduces the frequency of manual intervention, thereby enhancing the system's intelligence and operational stability.

[0097] This invention discloses an anti-blockage early warning and dynamic control method for underground belt conveyors in coal mines, effectively addressing the shortcomings of existing technologies, such as delayed early warning, passive control, and reliance on manual experience. Through multi-source heterogeneous data fusion and joint analysis, it achieves early and accurate quantitative identification of blockage risks. Based on multi-level early warning logic and adaptive control mechanisms, it significantly improves the timeliness of system response and the flexibility of handling strategies. This method transforms traditional experience-based operation and maintenance into intelligent proactive prevention and control, ensuring the continuous and stable operation of the underground transportation system while greatly reducing the risk of blockage accidents and improving the overall safety and reliability of operations.

[0098] Example 2 To achieve the above invention, embodiments of the present invention also provide a blockage prevention, early warning, and dynamic control system for underground belt conveyors in coal mines, such as... Figure 2 As shown, it includes: In one embodiment of the present invention, the overall system architecture design (a four-layer architecture of "perception, control, early warning, and decision-making") is as follows: Figure 2 (As shown).

[0099] Specifically, the sensing layer includes a multi-source sensor array deployment and data acquisition mechanism. Sensor types and deployment locations are as follows: Weight sensors: Resistance strain gauge weight sensors (model: YZC133) with a range of 0-500 kg / m and an accuracy class of 0.1%FS are installed on roller supports 10m upstream of the conveyor head discharge port and every 50m along the middle section, for a total of 35 sensors (adjusted according to belt length). These sensors are used to collect real-time material load values ​​per unit length of the belt (denoted as F, unit: kg / m). Speed ​​sensors: Hall effect speed sensors (model: HGC1030) with a measurement range of 0.2 m / s and a resolution of 0.01 m / s are installed on the tail roller shaft end. The sensors detect… The drum rotation speed is converted to the belt running speed (denoted as v, unit: m / s); Image sensor: a mine explosion-proof high-definition camera (model: KBA127) with a resolution of 1920×1080, a frame rate of 25fps, and a dustproof and waterproof rating of IP68 is used. It is installed 50cm above the head chute and on the top of the roadway in the middle section where material is prone to blockage. The lens faces the material conveying surface of the belt and is used to collect images of material accumulation; Data acquisition frequency: the sampling frequency of the weight sensor and speed sensor is set to 1 time / second, and the sampling frequency of the image sensor is set to 5 frames / second. All sensor data are transmitted to the local control unit via RS485 bus with a transmission rate of 115200bps and a data transmission delay of ≤100ms.

[0100] Specifically, the control layer (such as Figure 3 As shown): Local and Remote Dual-Mode Control Unit Design Local Control Unit (LCU): Hardware Configuration: Employs a mining explosion-proof PLC as the core controller, integrating an analog input module (SM1233), a digital output module (SM1222), and a fiber optic communication module, installed in the control chamber next to the underground belt conveyor; Core Functions: Receives data from the sensing layer, executes local control logic, sends speed adjustment commands to the belt motor frequency converter (model: MM440), and simultaneously establishes a bidirectional data link with the ground remote control platform through the fiber optic communication module; Ground Remote Control Platform (GRCP): Hardware Configuration: Consists of an industrial control computer (CPU i7 12700K, 32GB memory, 1TB SSD), a data storage server (using a RAID5 array, 16TB storage capacity), and a visual operation terminal; Software Functions: Develops a control interface based on LabVIEW, supporting real-time data display, historical data query, remote control command issuance, and early warning information management. The fiber optic communication bandwidth between the platform and the local control unit is ≥100Mbps, and the command transmission delay is ≤500ms.

[0101] Specifically, the early warning layer consists of a multi-source data fusion algorithm and a hierarchical early warning model. The load-speed matching algorithm determines whether the belt conveyor is in a "blockage risk zone," establishing the following formula for calculating the load-speed matching degree: M = (F × v) / (F0 × v0), where: M is the load-speed matching degree, reflecting the degree of matching between the current transport volume and the belt's rated transport capacity; F is the real-time load (kg / m), collected by a weight sensor; v is the real-time speed (m / s), collected by a speed sensor; F0 is the belt's rated load (kg / m), set according to the belt model (in this embodiment, F0 = 300 kg / m); v0 is the belt's rated speed (m / s), in this embodiment, v0 = 1.2 m / s. When M > 0.8, it is determined to be a "high-risk state," triggering an early warning (M > 0.8 indicates that the current transport volume exceeds 80% of the rated transport capacity, easily leading to blockage); when 0.5 ≤ M ≤ 0.8, it is determined to be a "normal state"; when M < 0.5, it is determined to be a "low-load state," triggering a light-load speed adjustment command.

[0102] Furthermore, the AI ​​image recognition algorithm (material accumulation detection based on YOLOv5) uses an improved YOLOv5 algorithm to process the material images acquired by the image sensor, as follows: Specifically, image preprocessing involves converting the acquired images to grayscale (formula: Gray = 0.299R + 0.587G + 0.114B, where R, G, and B are the pixel values ​​of the three image channels) and removing coal dust interference noise using Gaussian filtering (kernelsize = 3×3, σ = 1.0); dataset construction involves collecting 10,000 images of conveyor belt materials with different coal dust concentrations and particle sizes underground, labeling them with three categories: "material area," "conveyor belt edge," and "accumulation area," and constructing a training dataset; and model training is based on PyT. The YOLOv5 model was trained using the Orch framework. The input image size was set to 640×640, the batch size was 16, the training epochs were 100, the Adam optimizer was used, and the initial learning rate was 0.001. Stacking detection: The model outputs the bounding box of the material stacking area and calculates the ratio r = h / H of the stacking area height (h, in cm) to the belt side height (H, in cm). When r > 2 / 3, it is judged as "material stacking exceeds the limit" and an early warning is triggered. 1.3.3 The graded early warning mechanism is based on the joint judgment results of "load velocity matching degree M" and "image stacking ratio r", and the early warning level is divided into three levels: Level 1 warning (low risk): M>0.8 and r≤2 / 3, or r>2 / 3 and M≤0.8. At this time, only local audible and visual warning is triggered (sound frequency 800Hz, light flashing yellow, flashing frequency 1 time / second); Level 2 warning (medium risk): M>0.8 and r>2 / 3. At this time, local audible and visual warning is triggered (sound frequency 1000Hz, light flashing orange, flashing frequency 2 times / second), and warning information is pushed to the ground platform at the same time; Level 3 warning (high risk): If M is still>0.8 and r is still>2 / 3 within 10 seconds after the Level 2 warning is triggered, local audible and visual strong warning is triggered (sound frequency 1200Hz, light is solid red). The ground platform automatically pops up a window to prompt and sends an SMS to the mobile phone of the management personnel.

[0103] Specifically, at the decision-making level: adaptive control and remote handling logic. Adaptive speed control logic: Low-load speed regulation: When M < 0.5, the local control unit sends a speed adjustment command to the frequency converter, reducing the belt speed from the rated speed v0 to v1 = v0 × M (e.g., when M = 0.4, v1 = 1.2 × 0.4 = 0.48 m / s) to reduce motor energy consumption; High-risk speed regulation: When a first-level or second-level warning is triggered, the local control unit automatically reduces the belt speed from v0 to v2 = 0.8 m / s (experiments have verified that at this speed, material evacuation efficiency is improved by 40%, and the risk of material blockage is reduced by 70%), while continuously collecting the values ​​of M and r; Restoring normal speed: When M ≤ 0.8 and r ≤ 2 / 3 for 5 seconds, the local control unit sends a command to the frequency converter to gradually restore the speed from v2 to v0 (recovery rate 0.1 m / s, to avoid sudden speed increases causing material deviation).

[0104] Furthermore, the remote handling mechanism includes: Remote command issuance: When a Level 3 warning is triggered, the ground platform displays three operation options: "Remote shutdown," "Adjust unloading rate," and "Manual speed adjustment." After the manager selects an option, the command is transmitted to the local control unit via fiber optic cable, with an execution delay of ≤1 second; Unloading rate adjustment: If "Adjust unloading rate" is selected, the local control unit sends a command to the hydraulic valve of the unloading device at the machine head to adjust the opening of the unloading port from the rated opening (80%) to 60%, reducing the unloading amount per unit time. The rated opening is restored after M≤0.8; Emergency shutdown control: If "Remote shutdown" is selected, the local control unit first cuts off the power supply to the belt motor, then shuts down the unloading device, and simultaneously sends feedback on the shutdown status to the ground platform (feedback delay ≤500ms).

[0105] In one embodiment of the present invention, an auxiliary function module is designed.

[0106] Specifically, the local control unit of the sensor fault self-diagnosis module monitors the communication status and data validity of each sensor in real time. The diagnostic logic is as follows: Communication fault: If no data is received from a sensor for 5 consecutive seconds, it is determined to be a "communication fault". The system automatically switches to the backup sensor (each weight sensor and image sensor is equipped with one backup sensor). At the same time, a "sensor fault alarm" message is sent to the ground platform, marking the location of the faulty sensor (e.g., "middle section No. 2 weight sensor"). Invalid data: If the data collected by a sensor exceeds the range (e.g., weight sensor data > 500 kg / m or < 0 kg / m), it is determined to be "invalid data". The average value of the valid data in the previous 5 seconds is used instead, and the data is marked as "abnormal data" and stored in the historical database.

[0107] Specifically, the ground platform's data storage server for the historical data management and analysis module stores all data in the format of "timestamp, sensor type, data value, and warning status," with a storage period of ≥1 year. It supports the following functions: Data query: Retrieve historical data by time range, sensor type, and warning level, and generate Excel-format reports; Trend analysis: Automatically plot the "load time," "speed time," and "warning frequency time" curves for the past 30 days, using a moving average method to smooth data fluctuations; Risk assessment: Generate a monthly blockage risk analysis report, calculate the number of warnings and average handling time for each period, identify high-incidence periods and areas of blockage, and propose sensor optimization installation suggestions.

[0108] This invention discloses a blockage prevention, early warning, and dynamic control system for underground conveyor belts in coal mines. By constructing a four-layer collaborative architecture of "perception-control-early warning-decision," it effectively overcomes the fundamental shortcomings of traditional systems, such as delayed early warning, reliance on manual control, and lack of coordinated response. This system achieves closed-loop automated control throughout the entire process, from synchronous multi-source data acquisition, risk fusion and identification, intelligent hierarchical early warning, to adaptive speed adjustment and remote collaborative handling. It significantly improves the accuracy of blockage risk identification, the timeliness of early warning response, and the flexibility of control strategies. While ensuring the continuous, stable, and efficient operation of the underground transportation system, it greatly enhances its inherent safety level and intelligent operation and maintenance capabilities.

[0109] Example 3 To achieve the above invention, such as Figure 4 As shown, this embodiment also provides a coal mine underground belt conveyor anti-blockage early warning and dynamic control device 10, which includes: The multi-source sensor deployment and data synchronization acquisition module 100 is used to deploy a multi-source sensor array and synchronously acquire real-time load data, running speed data and material accumulation image data of the belt conveyor.

[0110] The dynamic matching degree calculation and image joint analysis module 200 is used to calculate the dynamic matching degree of load and speed based on real-time load data and running speed data, and to perform joint analysis by combining the material accumulation pattern recognition results output by the image recognition algorithm.

[0111] The multi-level early warning signal generation and triggering module 300 is used to generate early warning signals of corresponding early warning levels and trigger corresponding early warning devices based on the joint analysis results of dynamic matching degree and material accumulation pattern recognition results through multi-level early warning logic.

[0112] The local speed regulation and remote collaborative control module 400 is used to perform local adaptive speed regulation control and remote collaborative handling control according to the warning level. The local adaptive speed regulation control adjusts the belt running speed according to the dynamic matching degree, and the remote collaborative handling control receives and executes the handling instructions from the ground platform through the fiber optic communication link.

[0113] In one embodiment of the present invention, it further includes: a sensor fault self-diagnosis module, which is used to monitor the communication status and data validity of each sensor in real time through the sensor fault self-diagnosis module. If no sensor data is received for 5 consecutive seconds, it is determined to be a communication fault, automatically switches to the backup sensor, and sends sensor fault status information to the ground platform.

[0114] This invention discloses an anti-blockage early warning and dynamic control device for underground conveyor belts in coal mines. Through the collaborative operation of multiple modules, it effectively solves the core problems of existing technologies, such as single monitoring dimensions, delayed early warning response, and disconnect between local and remote control. The device achieves closed-loop management throughout the entire process, from synchronous acquisition, fusion analysis, and risk quantification of multi-source data to intelligent hierarchical early warning and adaptive collaborative control. This significantly improves the system's early identification accuracy of material blockage risks, the real-time performance of early warning responses, and the flexibility of control strategy execution. While reducing reliance on manual labor, it comprehensively enhances the proactive safety, process reliability, and intelligent management level of the underground transportation system.

[0115] To implement the methods of the above embodiments, the present invention also provides a computer device, such as... Figure 5 As shown, the computer device 600 includes a memory 601 and a processor 602; wherein, the processor 602 reads the executable program code stored in the memory 601 to run a program corresponding to the executable program code, so as to implement the various steps of the above-described method for anti-blocking early warning and dynamic control of a coal mine underground belt conveyor.

[0116] To implement the above embodiments, this application also proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a method for anti-blockage early warning and dynamic control of a coal mine underground belt conveyor as described in the foregoing embodiments.

[0117] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0118] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for anti-blockage early warning and dynamic control of underground belt conveyors in coal mines, characterized in that, include: S1, deploy a multi-source sensor array and synchronously collect real-time load data, running speed data and material accumulation image data of the belt conveyor; S2 calculates the dynamic matching degree of load and speed based on real-time load data and running speed data, and performs joint analysis by combining the material accumulation pattern recognition results output by the image recognition algorithm. S3, based on the joint analysis results of dynamic matching degree and material accumulation pattern recognition, generates warning signals of corresponding warning levels through multi-level warning logic and triggers corresponding warning devices. S4 executes local adaptive speed control and remote collaborative handling control according to the warning level; among them, local adaptive speed control adjusts the belt running speed according to the dynamic matching degree, and remote collaborative handling control receives and executes the handling instructions from the ground platform through the fiber optic communication link.

2. The method as described in claim 1, characterized in that, The deployment of a multi-source sensor array to synchronously collect real-time load data, operating speed data, and material accumulation image data of the belt conveyor includes: S11, Install the resistance strain gauge weight sensor on the idler roller bracket 10m upstream of the unloading port of the belt conveyor head and every 50m in the middle section, and set the sampling frequency to 1 time / second; S12, install the explosion-proof high-definition camera for mining at 50cm above the head chute and at the top of the roadway in the middle section where material is prone to blockage. The lens is facing the conveyor belt material conveyor surface, and the image sampling frequency is set to 5 frames / second.

3. The method as described in claim 1, characterized in that, The dynamic matching degree between load and speed is calculated based on real-time load data and operating speed data, including: S21, using formula Calculate the dynamic matching degree; where, As the reference load, As the reference speed; S22, performs grayscale conversion on the image data using the formula. The grayscale value is calculated, and Gaussian filtering is used to remove coal dust interference noise; where R, G, and B are the pixel values ​​of the three channels of the image.

4. The method as described in claim 1, characterized in that, Early warning signals are generated based on the joint analysis of dynamic matching degree and material accumulation pattern recognition results, including: S31, when the dynamic matching degree And image recognition results , and When the alarm is triggered, a level one warning is activated and the sound and light alarm is controlled to emit a sound signal at a frequency of 800Hz, while the yellow warning light flashes at a frequency of 1 time per second. S32, when the dynamic matching degree and When the alarm is triggered, a level-two warning is activated and the audible and visual alarm is activated to emit a sound signal at a frequency of 1000Hz. At the same time, the orange warning light flashes at a frequency of 2 times per second, and the warning information is pushed to the ground platform.

5. The method as described in claim 1, characterized in that, Local adaptive speed control is implemented based on the warning level, including: S41, when the dynamic matching degree At that time, a speed adjustment command is sent to the frequency converter to reduce the belt speed from the rated speed. Down to ; S42, when the first or second level warning is triggered, the belt speed is reduced from... Down to It gradually recovers to its original value at a gradient of 0.1 m / s. .

6. The method as described in claim 1, characterized in that, Also includes: S5 monitors the communication status and data validity of each sensor in real time through the sensor fault self-diagnosis module. If no sensor data is received for 5 consecutive seconds, it is determined to be a communication fault, automatically switches to the backup sensor, and sends sensor fault status information to the ground platform.

7. A device for anti-blockage early warning and dynamic control of underground belt conveyors in coal mines, characterized in that, include: The multi-source sensor deployment and data synchronization acquisition module is used to deploy a multi-source sensor array and synchronously acquire real-time load data, running speed data, and material accumulation image data of the belt conveyor. The dynamic matching degree calculation and image joint analysis module is used to calculate the dynamic matching degree of load and speed based on real-time load data and running speed data, and to perform joint analysis by combining the material accumulation pattern recognition results output by the image recognition algorithm. The multi-level early warning signal generation and triggering module is used to generate early warning signals of corresponding early warning levels and trigger corresponding early warning devices based on the joint analysis results of dynamic matching degree and material accumulation pattern recognition results through multi-level early warning logic. The local speed regulation and remote collaborative control module is used to perform local adaptive speed regulation control and remote collaborative handling control according to the warning level. The local adaptive speed regulation control adjusts the belt running speed according to the dynamic matching degree, and the remote collaborative handling control receives and executes the handling instructions from the ground platform through the fiber optic communication link.

8. The apparatus as claimed in claim 7, characterized in that, Also includes: The sensor fault self-diagnosis module is used to monitor the communication status and data validity of each sensor in real time. If no sensor data is received for 5 consecutive seconds, it is determined to be a communication fault, automatically switches to the backup sensor, and sends sensor fault status information to the ground platform.

9. An electronic device, comprising: processor; The memory stores executable instructions; when the processor executes the instructions, it implements the anti-blocking early warning and dynamic control method for underground belt conveyors in coal mines as described in any one of claims 1-6.

10. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements a method for anti-blockage early warning and dynamic control of a coal mine underground belt conveyor as described in any one of claims 1-6.

Images